The announcement positions AirPods 5 as a value-focused upgrade rather than a complete redesign. The familiar white earbud shape remains, but the changes inside are meant to improve sound, comfort, and everyday usability. Apple says the new model delivers 50% more noise reduction than AirPods 4, a claim that would make the base AirPods far more effective in noisy environments than previous entry-level versions.

Key facts at a glance

• AirPods 5 were announced at Apple's iPhone 18 Pro event.
• The earbuds offer 50% more noise reduction compared with AirPods 4.
• Apple redesigned the internals and added next-generation Adaptive EQ.
• Personalized Spatial Audio is said to sound better than before.
• Battery life reaches up to 20 hours with the charging case.
• Two versions are available, and both include Active Noise Cancellation.
• The higher-end model adds on-stem volume controls, longer battery life, and a wireless charging case.
• Pricing starts at $129, with the higher-end model at $149.
• Pre-orders begin today, and units ship Friday, September 18.

A major ANC upgrade for the entry-level line

Active noise cancellation has become one of the most important features for wireless earbud buyers. It helps reduce the low-frequency rumble of airplanes, trains, and city traffic, and it can make music, podcasts, and calls easier to hear in crowded places. For years, Apple reserved its best ANC performance for the AirPods Pro line, which uses silicone ear tips to create a seal inside the ear canal. The standard AirPods, by contrast, have an open-fit design that lets in more ambient sound.

With AirPods 4, Apple finally brought ANC to the standard AirPods family, but only on the higher-end version. The base model remained without noise cancellation. AirPods 5 changes that approach. Both versions now include Active Noise Cancellation support, which means buyers no longer have to choose the more expensive tier just to get basic noise reduction. That shift could make the new entry-level model more appealing to commuters, students, and travelers who want ANC without paying Pro-level prices.

The claim of 50% more noise reduction than AirPods 4 is notable because it suggests Apple has improved both the microphones and the processing algorithms. Noise cancellation depends on a combination of external microphones, internal microphones, and real-time audio processing. The system listens to the outside world, analyzes unwanted noise, and generates an opposite sound wave to cancel it out. Better microphones and faster processing can make that cancellation more effective, especially across a wider range of frequencies.

However, open-fit earbuds still face physical limits. Because they do not seal the ear canal, they cannot block as much sound passively as in-ear models with silicone tips. Even with improved ANC, AirPods 5 may not match the isolation of AirPods Pro 3 or other sealed earbuds. For many users, though, the convenience of an open-fit design is worth the trade-off, and the improved ANC should make the new model more usable in noisy settings than any previous standard AirPods.

Redesigned internals and next-generation Adaptive EQ

Apple says AirPods 5 feature redesigned internals. That phrase can cover a lot of ground, but in practice it usually points to changes in the acoustic chamber, drivers, microphones, and ventilation. A redesigned interior can improve bass response, reduce distortion, and make the earbuds more comfortable over long listening sessions. It can also create more space for additional sensors and batteries, which is important as Apple adds more features to a small device.

The next-generation Adaptive EQ is another key upgrade. Adaptive EQ uses microphones to monitor what the listener hears and adjusts the low and mid frequencies in real time. The goal is to deliver consistent sound quality regardless of how the earbuds fit or how they move during activity. If the seal changes slightly, or if one earbud sits differently than the other, Adaptive EQ can compensate. The new version is said to be more advanced than the system used in AirPods 4, which should mean more natural vocals, clearer instruments, and more balanced bass.

Personalized Spatial Audio is also getting an update. Spatial Audio creates a surround-sound-like experience by placing sound around the listener rather than simply inside their head. Personalized Spatial Audio takes that idea further by using a scan of the user's ears to create a custom profile. That profile helps the system place sounds more accurately for that specific person. Apple says the feature sounds better on AirPods 5, which could make movies, games, and live recordings more immersive.

Battery life and everyday use

Battery life remains a critical factor for wireless earbuds. AirPods 5 offer up to 20 hours of listening time with the charging case, according to Apple. That figure includes multiple charges from the case, so users can top up throughout the day without finding a wall outlet. The higher-end model adds longer battery life on top of that, although Apple has not framed the difference as a dramatic leap. For most people, the standard model's 20-hour total should be enough for a full day of commuting, work, and exercise, especially with short breaks in the case.

The charging case itself is an important part of the experience. The standard AirPods 5 come with a case that supports wired charging, while the higher-end version includes a wireless charging case. Wireless charging is convenient for users who already have a Qi charging pad on a desk or nightstand. It also reduces the need to carry a cable, which is a small but meaningful quality-of-life improvement for frequent travelers.

Two models, both with ANC

Apple is selling AirPods 5 in two versions. The main difference is not noise cancellation, because both models support ANC. Instead, the higher-end version adds features that were previously reserved for more expensive earbuds. These include on-stem volume controls, similar to the controls found on AirPods Pro. That may sound like a minor detail, but on-stem volume controls are genuinely useful. They let users adjust volume without reaching for their phone, asking Siri, or using a connected device. It is the kind of feature that makes daily use smoother.

The higher-end AirPods 5 also offer longer battery life and a wireless charging case. Together, those additions justify the $20 price difference for many buyers. The standard model at $129 is still a strong value, especially now that ANC is included, but the $149 model may be the better choice for people who want the most convenient experience.

The two-tier strategy mirrors how Apple has handled other product lines. It gives budget-conscious buyers a clear entry point while offering a step-up option for those who want extra features. The fact that both tiers now include ANC simplifies the decision. Shoppers no longer have to give up noise cancellation to save money, which was the case with the AirPods 4 lineup.

Pricing and availability

AirPods 5 start at $129. That is the same starting price as AirPods 4, but the new model includes ANC at that price. The higher-end AirPods 5 with the wireless charging case cost $149. Pre-orders open today, and the first orders are scheduled to ship on Friday, September 18.

The pricing is aggressive. At $129, AirPods 5 compete directly with a growing number of noise-canceling earbuds from other brands. Many of those rivals offer ANC, but they often lack the deep integration Apple provides across iPhone, iPad, Mac, and Apple Watch. Features like quick pairing, automatic switching, Find My support, audio sharing, and hands-free Siri make AirPods convenient for people already inside Apple's ecosystem. The lower price and improved ANC could convince more iPhone owners to upgrade from older AirPods or to choose AirPods over third-party alternatives.

What the update means for the AirPods lineup

AirPods 5 occupies an interesting position in Apple's audio lineup. It sits below AirPods Pro 3, which remains the choice for users who want the best noise cancellation, a sealed fit, and advanced features like transparency mode and conversation awareness. It also sits below AirPods Max 2, the over-ear model for listeners who prioritize size, comfort, and high-fidelity sound. By adding ANC to both AirPods 5 versions, Apple has made the standard AirPods more competitive without directly replacing the Pro or Max models.

The gap between AirPods 5 and AirPods Pro 3 may still matter. AirPods Pro 3 have silicone tips, which create a seal and improve both passive and active noise cancellation. They also offer features such as adaptive transparency, which lets outside sound in when needed, and they are often discounted during major sales events. A higher-end AirPods 5 at $149 could tempt buyers who want on-stem controls and wireless charging but do not want to spend $199 or more on Pro earbuds. That could put some pressure on Apple's own premium models, but it also gives the company a stronger answer to competitors at the $129 to $149 price point.

For current AirPods 4 owners, the upgrade question comes down to how much they value noise cancellation. If they already own the higher-end AirPods 4 with ANC, the improvements in AirPods 5 may be incremental rather than revolutionary. If they own the base AirPods 4 without ANC, or an even older model, the jump to AirPods 5 could feel much more significant. The combination of ANC, redesigned internals, Adaptive EQ, better Personalized Spatial Audio, and up to 20 hours of battery life with the case addresses several common complaints about entry-level wireless earbuds.

Accessories and current deals to consider

Buyers who are considering AirPods 5 may also look at accessories and deals on other models. The AirPods Pro 3 are currently available for $199, down from $249. The AirPods Max 2 are priced at $449, down from $549. Both discounts make Apple's higher-end audio products more accessible during the launch window.

For everyday care, a cleaning kit for AirPods can help remove earwax and debris from the speaker mesh and microphone ports. A magnetic strap neckband can prevent earbuds from being lost when they are not in the case. The AirFly Pro 2 is an adapter that connects AirPods to standard headphone jacks, which is useful on airplanes and in gyms that still use older audio systems. These accessories do not change the core AirPods 5 experience, but they can make the earbuds more practical in specific situations.

Competitive pressure and the road ahead

The wireless earbud market is crowded. Samsung, Google, Sony, Bose, and many smaller brands all sell noise-canceling models at a variety of prices. Apple's advantage has always been integration, but integration alone is not enough when competitors offer strong ANC and long battery life for less. By lowering the barrier to ANC and keeping the starting price at $129, Apple is making a clear play for value-conscious buyers who still want a polished experience.

The success of AirPods 5 will depend on real-world performance. Apple's 50% noise reduction claim will need to hold up in tests, especially in environments with continuous low-frequency noise like airplane cabins and train stations. The redesigned internals and Adaptive EQ will also need to deliver noticeable improvements in sound quality. If they do, AirPods 5 could become the default recommendation for iPhone users who want wireless earbuds without paying Pro prices. If the improvements are subtle, the higher-end model's extra features may be the main reason to choose it over the standard version.

Apple has not announced a specific end date for pre-orders, but the shipping date gives buyers a clear timeline. The first AirPods 5 units are set to arrive on Friday, September 18.

Source: 9to5Mac News

Design and display changes

The iPhone 18 Pro and iPhone 18 Pro Max retain the familiar Pro silhouette, but Apple is giving buyers more finish choices. The phones will be available in black, glacier blue, silver, and burgundy red. The burgundy option is a particularly notable addition because it brings a deeper, warmer tone to the Pro range, while glacier blue offers a lighter, cooler alternative. The company says the new models feature a new approach to the back design, though it has not abandoned the overall shape that has defined recent Pro models.

One of the most visible changes is the Dynamic Island. Apple has made it smaller on both the iPhone 18 Pro and iPhone 18 Pro Max, and it can now display three Live Activities at once. That means users can keep tabs on more real-time information without opening separate apps. The expanded Live Activity support could be useful for sports scores, delivery tracking, travel updates, timers, and other ongoing tasks. By shrinking the cutout while increasing its information density, Apple is trying to make the front of the phone feel more immersive without removing the interactive area entirely.

A20 Pro chip and AI ambitions

At the heart of the iPhone 18 Pro is the A20 Pro chip, built on a 2-nanometer process. The move to 2nm is significant because it allows Apple to pack more transistors into a smaller area while improving power efficiency. The A20 Pro features a 6-core CPU with two super cores and four efficiency cores. That configuration is designed to handle demanding workloads quickly while conserving battery during lighter tasks.

Graphics performance also gets a boost. The chip includes a 7-core GPU that Apple says delivers 40 percent faster graphics performance. For gamers, video editors, and users who rely on graphically intensive apps, that improvement could translate into smoother frame rates and faster rendering. The Neural Engine has been expanded to 32 cores, and Apple is adding a second Neural Engine. The company is framing the iPhone 18 Pro as the ultimate AI device, a clear signal that on-device machine learning, generative features, and intelligent assistants are central to its strategy.

Memory bandwidth is another area of improvement. Apple claims a 50 percent increase in memory bandwidth, which should help the CPU, GPU, and Neural Engine move data more efficiently. Higher bandwidth is especially important for AI models, high-resolution photo and video processing, and multitasking. The A20 Pro is also directly connected to a vapor chamber for more effective thermal management. This direct connection is intended to move heat away from the chip more efficiently, reducing the chance of throttling during extended gaming, video export, or AI processing sessions.

Thermals and battery life

Apple is emphasizing sustained performance with the iPhone 18 Pro and iPhone 18 Pro Max. A next-generation vapor chamber helps deliver that sustained performance, meaning the phones should maintain higher speeds for longer periods instead of dropping performance when temperatures rise. This matters for professionals who use their phones for heavy tasks and for anyone who has experienced a device slowing down during intense use.

Battery life is another headline feature. Apple says the iPhone 18 Pro matches the battery life of the iPhone 17 Pro Max, while the iPhone 18 Pro Max delivers up to 30 hours of battery life. That is a notable claim for the larger model, and it suggests that the combination of the 2nm A20 Pro chip, improved thermal design, and software optimization is paying off. Wired charging is also faster, which should reduce the time users spend tethered to a power source. Faster charging is increasingly important as phones take on more demanding roles, from navigation and streaming to AI-powered tools.

Camera upgrades and Pro controls

The camera system receives several meaningful upgrades. The iPhone 18 Pro includes a 48MP Fusion main camera with variable aperture, and this feature is not exclusive to the Pro Max. Variable aperture allows the camera to adjust how much light reaches the sensor. A wider aperture can help in low light, while a narrower aperture can increase depth of field so more of a scene stays in focus. Apple says night mode captures are now faster and better in quality as a result.

Pro controls have been expanded as well. The Camera app now offers a customizable interface with new options for manual control. Users can manually adjust aperture and shutter speed, giving photographers more creative flexibility. A smart focus mode improves locking onto a subject, which should help with moving people, pets, and action shots. Photographic styles are also more advanced, with texture controls and other adjustments that let users fine-tune the look of their images without losing the original detail.

Apple is also introducing a feature called Apple Reference Image. This system proves the authenticity of a photo by adding metadata that ensures an image is not AI-generated. It stores a reference image alongside image edits so the source material can always be verified. As AI-generated imagery becomes more common, tools that establish provenance could become increasingly important for journalists, photographers, and everyday users who want to confirm that a picture is real.

Pricing, storage, and availability

The iPhone 18 Pro starts at $1199, and it includes a 2TB option for the first time. That new storage tier gives power users more room for high-resolution photos, 4K and 8K video, large apps, and offline media. The iPhone 18 Pro Max starts at $1299, which is a $100 price increase over the 17 Pro models. The higher starting price reflects the advanced components, the larger battery, and the broader feature set, though it also places the Pro Max further into premium territory.

Apple is releasing new accessories alongside the phones. These include an updated Clear Case design, a wrist strap, and new colors of silicone and FineWoven cases. The accessories are designed to complement the new color options and give users more ways to carry and protect their devices. AppleCare One Family plan is also launching in the United States, offering coverage for multiple Apple products under a single plan.

Pre-orders for the iPhone 18 Pro and iPhone 18 Pro Max begin on September 12, and the official launch is set for September 18. That gives buyers a short window between announcement and availability, which is typical for Apple's major iPhone releases.

What the upgrades mean for the Pro lineup

The iPhone 18 Pro and Pro Max arrive at a time when smartphone makers are competing heavily on AI features, camera quality, and battery life. Apple's decision to build the A20 Pro on a 2nm process and pair it with a second Neural Engine shows that the company sees on-device AI as a defining feature of the next generation. The 32-core Neural Engine and 50 percent increase in memory bandwidth are not just spec sheet numbers; they are meant to support real-time language translation, image generation, advanced photo editing, and smarter assistants.

The variable aperture camera is another sign that Apple is pushing computational photography further. While variable aperture has appeared in some dedicated cameras and select smartphones, bringing it to both Pro models suggests that Apple wants to give more users manual control over depth of field and exposure. Combined with the new Pro controls and smart focus mode, the iPhone 18 Pro is being positioned as a tool for photographers who want more than automatic point-and-shoot results.

Battery life and thermal management may be less glamorous than camera specs, but they are often what separate a good phone from a great one. The next-generation vapor chamber and the direct connection between the A20 Pro and the cooling system indicate that Apple is addressing sustained performance. If the iPhone 18 Pro Max can truly deliver up to 30 hours of battery life while handling demanding tasks, it could appeal to travelers, creators, and professionals who need their phone to last through long days.

The pricing strategy is also worth noting. The iPhone 18 Pro starts at $1199, while the Pro Max starts at $1299. The $100 increase over the previous Pro models may give some buyers pause, but the addition of a 2TB option and the expanded feature set could justify the cost for users who need maximum storage and performance. The four color options, including burgundy and glacier blue, give the lineup more personality at a time when many phones look similar from the front.

With pre-orders opening on September 12 and the launch on September 18, the iPhone 18 Pro and iPhone 18 Pro Max are set to become Apple's most advanced smartphones. The combination of the A20 Pro chip, variable aperture camera, smaller Dynamic Island, improved battery life, and new accessories makes this a significant update for the Pro line, even as it retains the design language of the previous generation.

Source: 9to5Mac News

The headline feature is the display system. The inner panel measures 7.6 inches and uses Super Retina XDR technology, while the outer display measures 5.4 inches. Apple has applied a custom nano-texture finish that is intended to minimize crease visibility and reduce glare. The company also developed an Apple-designed coating for scratch resistance. A FaceTime camera is hidden under the inner display, which helps preserve an uninterrupted viewing area. The outer screen includes a 12MP Center Stage front camera for selfies and video calls. Together, the two displays allow users to move between quick phone tasks and a larger canvas for multitasking, media, and productivity.

Design, materials, and durability

The iPhone Duo is made from grade 5 titanium, a material Apple has increasingly used across its premium products. It will be available in star white and night sky colors. The frame is designed to balance rigidity with low weight, while the hinge and folding mechanism are said to have been tested for repeated use. Apple rates the device IP68 for dust and water resistance, a significant durability claim for a foldable. Foldables have historically faced questions about hinge reliability, display creasing, and protection from dust and water. By offering IP68, Apple is signaling that the iPhone Duo can handle everyday exposure to splashes, rain, and dust without requiring a radically different level of care than a standard iPhone.

Apple also says the iPhone Duo can be used in any orientation or pose. The iOS software intelligently reacts to the position of the device, whether it is open flat, partially folded, or closed. That flexibility is central to the foldable experience. A partially folded pose can turn the lower half into a stand or a control surface, while the upper half can display content. The device also supports Split View multitasking, allowing users to pin a video to the top of the display and keep using the other half for messaging, browsing, or note-taking. The result is a device that adapts its interface to the shape it is in rather than forcing users into a single posture.

Touch ID and Apple Watch unlock

For biometric authentication, Apple has integrated Touch ID into the side button. Users can enroll multiple fingers, which makes it easier to unlock the device with either hand or while the phone is folded. Apple also supports unlocking the iPhone Duo with an Apple Watch. The combination of side-button Touch ID and Apple Watch unlock gives users more options depending on how they are holding or wearing the device. This approach differs from Face ID-only iPhones, but it reflects the physical realities of a foldable. When the phone is closed, the outer display is small, and when it is open, the inner camera is under the display. A side-mounted fingerprint sensor provides a consistent authentication method across those states.

Performance and battery life

The iPhone Duo is powered by the A20 Pro chip, which includes a new display engine that drives both displays simultaneously. That engineering detail matters because foldables must manage two screens, different refresh behaviors, and seamless transitions without draining the battery or causing visual stutter. Apple has also included the C2 cellular modem, continuing its work on custom wireless components. Battery life is split across each side of the device. Apple claims 31 hours of video playback using the inner display and 44 hours using the outer display. The outer screen is smaller and likely more power-efficient, while the inner display offers a larger viewing area at a higher battery cost. The dual-battery design allows the two halves to distribute weight and power more evenly.

For audio, the iPhone Duo includes speakers on the top and bottom, enabling spatial audio. That layout should help when the device is open and being used for video, games, or calls. The company has also reimagined Standby for the iPhone Duo, suggesting that the always-on or glanceable information experience has been adapted to the foldable form factor. Standby could take advantage of the outer display when the phone is closed or the inner display when it is partially folded on a desk. Apple has not yet detailed every software feature, but the combination of a new display engine, spatial audio, and a reworked Standby mode indicates that the iPhone Duo is meant to be used in more contexts than a traditional phone.

Camera system

The camera setup includes a 48MP Fusion Main camera with 12MP 2x optical zoom and support for 4K video at 120fps. There is also a 48MP Fusion Ultra Wide camera. On the front, a 12MP Center Stage camera sits on the outer display, while a FaceTime camera is hidden under the inner display. The camera arrangement is designed to work whether the phone is closed, open, or partially folded. The outer display camera is likely the primary option for quick selfies and video calls when the device is shut. When open, the under-display FaceTime camera allows for larger video calls without a visible camera cutout. The main and ultra wide cameras provide the high-resolution stills and video that users expect from a modern iPhone flagship.

Apple has also confirmed Apple Pencil support coming later this year. That addition would make the iPhone Duo more than a foldable phone. With a 7.6-inch inner display, the device could serve as a compact sketchbook, note-taking tool, and document annotation device. Apple Pencil support would also help differentiate the iPhone Duo from other foldables that focus mainly on media consumption. The timing, coming later in the year, suggests that Apple wants to launch the phone first and then expand its capabilities through a subsequent update or accessory release. For creative users, the combination of a large inner display, titanium body, and stylus input could make the iPhone Duo a genuinely new category within the iPhone lineup.

Accessories and software experience

Apple is offering new accessories built for the foldable design. The iPhone Duo Folio includes a kickstand, and there is a two-part case that protects each half of the device. Those accessories acknowledge that a foldable phone has different protection needs than a standard slab phone. A two-part case must accommodate the hinge while still allowing the phone to open and close freely. The Folio with kickstand suggests that Apple expects users to prop the device up for video, video calls, or multitasking. The software experience appears to be equally important. Split View, orientation awareness, partial folding, and a reworked Standby mode all point to an interface that changes based on how the device is positioned. That is the core promise of a foldable: one device that can be a phone, a small tablet, and a hands-free display.

The iPhone Duo starts at $1,999 for the 256GB model. Preorders begin on October 16, and the device will be available on October 23. That pricing places the iPhone Duo at the top of Apple’s iPhone lineup, above even the most expensive Pro Max models. The price reflects the complexity of the foldable display, the titanium construction, the dual-battery system, and the custom engineering required to drive two screens. It also signals that Apple is treating the iPhone Duo as a premium product for early adopters and power users rather than a mass-market replacement for every iPhone. The October release window gives Apple time to build inventory and prepare software updates, including the promised Apple Pencil support later in the year.

What the iPhone Duo means for the foldable market

Apple’s entry into foldables has been rumored for years, and the iPhone Duo finally makes it official. The company is arriving later than some competitors, but it is doing so with a distinctive set of choices: a book-style fold, a titanium body, IP68 durability, side-button Touch ID, a hidden inner camera, and a focus on multitasking. The foldable market has matured over the past several years, with multiple manufacturers offering devices that open into tablet-sized screens. Apple’s version is likely to draw intense attention because of the size of the iPhone user base and the tight integration between iOS, Apple silicon, and services. If the iPhone Duo succeeds, it could

Source: 9to5Mac News

In a collaboration that brought together two distinct corners of internet culture, a football content creator spent a full day with Khaby Lame, the silent comedy star and one of the most-followed personalities on social media. The goal was simple but ambitious: to experiment with different ways to score, blending traditional football mastery with the kind of viral, common-sense moments that have made Khaby Lame a global phenomenon. The result was a session filled with laughter, trial and error, and some truly unique shots that showcased both technical skill and comedic timing.

The creator, known for detailed football tutorials and trick demonstrations, said the idea came from a belief that creativity has no limits. By stepping into Khaby Lame's universe, the session aimed to merge two worlds: the precision of technical football and the accessible humor of viral content. What emerged was not just a series of trickshots but a study in how different forms of entertainment can complement each other.

Who Is Khaby Lame?

Khaby Lame, born in Senegal and raised in Italy, rose to fame on TikTok with a style that requires no spoken language. His videos typically feature him silently reacting to overly complicated life hacks, offering a simple, common-sense solution with a signature shrug and facial expression. That universal appeal helped him become the most-followed creator on TikTok, with a following that exceeds 160 million people. His content has since expanded into broader entertainment, including brand partnerships, television appearances, and sports-related collaborations.

Before social media, Khaby Lame worked in a factory. After losing his job, he began posting short videos during the early days of the pandemic. His wordless humor quickly resonated across languages and cultures. The simplicity of his reactions became his trademark: while others overexplained, he simply showed how easy a task could be. That same ethos carried into the football trickshot session, where complexity was often met with a smile and a straightforward alternative.

Merging Football Mastery With Viral Humor

The session took place on a football pitch, with cameras rolling to capture every attempt. The creator brought a bag of technical challenges: volleys, rabonas, no-look passes, backheel finishes, and long-range strikes. Khaby Lame, initially known more for comedy than sport, quickly showed that he was not merely a participant. According to the creator, Khaby took to the technical side of the challenge with surprising naturalness, displaying a real eye for precision. That observation became one of the key takeaways from the day.

They began with basic drills to warm up. The creator demonstrated proper form for striking the ball, controlling the pace, and directing shots into corners. Khaby Lame watched closely, then attempted each technique. Early attempts were mixed, as expected, but his improvement was rapid. He started to adjust his body shape, angle his run-up, and connect with the ball more cleanly. The creator noted that Khaby's background in physical comedy may have helped: timing, coordination, and body awareness are central to both silent humor and football skills.

Traditional Technical Shots

The first phase focused on traditional technical shots. They set up targets in the goal and practiced different finishes. The creator emphasized fundamentals: plant foot next to the ball, strike through the center, follow through toward the target. Khaby Lame applied these instructions with increasing consistency. He attempted a curled shot into the top corner, a low driven effort, and a chipped finish over a dummy goalkeeper. Each attempt was captured for social media, but the emphasis remained on genuine improvement rather than rehearsed perfection.

One highlight came when Khaby tried a rabona. The rabona, a technique where the kicking leg wraps around the standing leg, is difficult even for experienced players. After a few awkward attempts, Khaby adjusted his approach and produced a clean strike that found the net. The creator was visibly impressed, noting that such precision usually takes years of practice. The moment underlined a theme of the day: Khaby Lame may be known for comedy, but he is also a quick learner with physical intelligence.

Viral Common-Sense Moments

The second phase introduced the viral, common-sense style that defines Khaby Lame's online persona. Instead of executing an overly complicated trick, he would sometimes simplify the challenge. For example, when the creator set up a sequence involving multiple stepovers, a fake shot, and a delicate chip, Khaby Lame watched, shrugged, and simply passed the ball into an open goal. The contrast between the elaborate setup and the simple finish was pure Khaby Lame. It made the point that creativity does not always require complexity.

These moments were not dismissals of technical skill. Rather, they were comedic counterpoints that highlighted the absurdity of overcomplication. The creator embraced the contrast, and the two soon developed a rhythm: one would propose a difficult trick, the other would find the simplest possible solution. The result was a series of clips that felt both athletic and accessible. Fans of football could appreciate the technique, while fans of Khaby Lame could enjoy the familiar humor.

Behind the Scenes of the Session

The day was not without its challenges. Weather conditions shifted, requiring adjustments to the schedule. Equipment needed to be reset after particularly powerful shots. The creator and Khaby Lame also had to manage the demands of filming, ensuring that each attempt was captured from multiple angles without disrupting the flow. Despite these logistical layers, the atmosphere remained light. Laughter was constant, and the two collaborators seemed to genuinely enjoy the process.

Between takes, they discussed the nature of viral content. The creator explained how football trickshots often require repeated attempts to get the perfect clip. Khaby Lame, with his background in short-form video, understood the value of patience and repetition. He also shared insights into his own creative process, noting that simplicity is often the hardest thing to achieve. That conversation added depth to the collaboration, moving it beyond a simple stunt and into a genuine exchange of creative perspectives.

They also experimented with different scoring methods that mixed football and everyday objects. In one sequence, they used a set of cones to create a mini obstacle course. In another, they challenged each other to score with a single touch. The variety kept the session fresh and allowed both participants to showcase different skills. The creator's technical knowledge and Khaby Lame's comedic instincts complemented each other, producing moments that were neither purely instructional nor purely humorous.

Why This Collaboration Matters

The intersection of sports and viral entertainment is not new, but the pairing of a football specialist with one of the world's most recognizable social media personalities is notable. It reflects a broader trend: audiences increasingly expect athletes and creators to cross traditional boundaries. Football clubs, players, and brands have all leaned into short-form content, recognizing that reach often comes from personality as much as performance. Khaby Lame's involvement adds a layer of mainstream appeal that extends far beyond football fandom.

For the creator, the session was an opportunity to introduce technical football concepts to a wider audience. Khaby Lame's followers may not tune in for tactical breakdowns, but they will watch for his reactions and humor. By embedding football skills within that framework, the collaboration could spark curiosity in new viewers. Some may try the tricks themselves. Others may simply enjoy the entertainment. Either way, the content achieves something valuable: it makes football feel approachable and fun.

For Khaby Lame, the session adds to a growing portfolio of sports-related content. He has previously appeared at football events, interacted with professional players, and attended high-profile matches. Those appearances have shown that his appeal translates beyond the phone screen. The trickshot session continues that trajectory, demonstrating that he can hold his own in a sporting context while remaining true to his comedic identity. The balance is delicate, but the day's footage suggests he found it naturally.

What Fans Can Expect

While the full set of clips has not yet been released, early descriptions suggest a mix of spectacular finishes and comedic moments. Fans can expect to see Khaby Lame attempting volleys, curling shots, and possibly a bicycle kick. They can also expect the signature shrug, the silent reaction, and the simple solutions that made him famous. The creator has hinted that some shots were genuinely surprising, with Khaby Lame exceeding expectations on more than one occasion.

The content is likely to be distributed across multiple platforms, with short clips optimized for vertical video and longer edits for dedicated football channels. This multi-platform approach is standard for high-profile collaborations, allowing each audience to engage in the format they prefer. The creator's existing followers will appreciate the technical details, while Khaby Lame's massive audience will bring new eyes to the football content. The combination could produce significant reach.

Beyond the numbers, the session represents a creative experiment. It asks what happens when two different approaches to entertainment share the same space. The answer, based on the day's events, is a productive tension between precision and simplicity. That tension is not a conflict but a dialogue. The creator pushes for technical excellence; Khaby Lame pulls toward accessible humor. The result is content that can be enjoyed on multiple levels.

Expanding the Universe of Trickshots

Trickshot videos have long been a staple of football content. From freestyle street football to studio-based challenges, creators have found ways to blend skill with spectacle. The genre thrives on innovation, as creators constantly seek new angles, new props, and new collaborators. Bringing in a personality like Khaby Lame expands the genre's possibilities. It suggests that trickshot content does not have to be serious to be impressive. Humor can coexist with skill.

The session also highlights the role of common sense in creativity. Khaby Lame's entire brand is built on the idea that many problems have simple solutions. In football, that idea translates to efficient play: the best finish is often the simplest one. A perfectly placed side-foot shot can be more effective than an elaborate trick. By repeatedly demonstrating this principle, Khaby Lame brought a refreshing perspective to a genre that sometimes prioritizes style over substance.

At the same time, the creator's willingness to embrace that perspective shows confidence in their own expertise. Rather than defending complexity, they allowed the contrast to play out. The result was a collaborative atmosphere where both participants learned from each other. The creator gained insight into comedic timing and audience engagement. Khaby Lame gained experience in football technique. The exchange was mutual, which is likely why the footage feels authentic.

Khaby Lame's Football Connections

Khaby Lame's interest in football is not new. He has attended matches, met players, and participated in charity games. His silent reactions have been applied to football moments, from penalty shootouts to skill challenges. In 2023, he appeared in a promotional video with a Serie A club, and he has been photographed with stars from the international game. Those appearances helped pave the way for the trickshot session. They also showed that his brand can adapt to sports without losing its core identity.

The football world has welcomed him because his appeal is broad and non-threatening. He does not mock the sport; he engages with it. His humor is gentle, focused on common sense rather than criticism. That makes him a safe and effective collaborator for teams, players, and creators. The trickshot session fits that pattern. It showcases football in a positive light while allowing Khaby Lame to do what he does best: react, simplify, and entertain.

The Creative Process

Creating a trickshot video requires planning, patience, and flexibility. The creator typically starts with a concept, then breaks it down into individual shots. Each shot is practiced multiple times before filming. During the session with Khaby Lame, the process was more improvisational. Because Khaby's reactions were unpredictable, the creator had to adapt on the fly. That spontaneity became an asset, leading to moments that could not have been scripted.

The cameras captured more than just the successful attempts. Misses, near-misses, and comedic failures were also recorded. Those outtakes are often valuable because they show the human side of the process. Audiences enjoy seeing effort and imperfection. For Khaby Lame, whose brand is built on authenticity, the unpolished moments were particularly on-brand. They demonstrated that even a global star can laugh at himself when a shot goes wide.

Editing will play a crucial role in shaping the final narrative. The goal is to balance technical showcases with comedic beats. Too much technique may alienate casual viewers; too much comedy may undermine the football content. The creator and Khaby Lame's team will likely work together to find the right mix. Based on the day's events, there is plenty of material to achieve that balance.

Looking Ahead

The collaboration may lead to further projects. Given the positive experience, both parties have expressed openness to future sessions. Possible ideas include a rematch with more challenging shots, a series of common-sense football tips, or a charity match featuring other creators. While nothing has been confirmed, the success of the initial day suggests that the partnership has potential beyond a single video. The football creator's audience and Khaby Lame's fanbase are both large, and their overlap is growing.

For now, the focus remains on the footage captured during the session. Editing will determine the final shape of the content, but the raw material is rich. There are moments of genuine skill, moments of comedy, and moments of unexpected synergy. The creator has described the day as filled with laughter and unique shots. That description aligns with what audiences have come to expect from both figures: entertainment that feels spontaneous rather than forced.

The session added another layer to Khaby Lame's growing catalogue of sports-themed content, which already includes appearances at football events and collaborations with professional players. It also reinforced the creator's reputation for pushing creative boundaries. By stepping into Khaby Lame's universe, the creator demonstrated that football mastery can be adapted to any format.

Source: MSN News

That was the state of my Google Drive: roughly 340GB of documents, spreadsheets, slide decks, PDFs, screenshots, tax filings, and downloads, most of it accumulated since the account was first created. Add nearly 40,000 photos sitting in a separate cloud photo library and a handful of overlapping subscriptions, and the monthly bill starts to feel less like a convenience and more like a tax on indecision.

The Economics of Digital Hoarding

The free tier of Google's consumer storage is 15GB, shared across Drive, Gmail, and Photos. For someone with hundreds of gigabytes of professionally generated documents, that ceiling is not an aspirational target. It is a fantasy. There is no plausible sequence of productive afternoons in which that data shrinks by 95 percent without a radical change in behavior or tooling.

So the subscriptions multiply. A 5TB Google AI plan runs about $20 a month and bundles Gemini access. A 2TB iCloud plan adds another $10. A separate AI chatbot subscription tacks on $20 more. Individually, each charge is defensible. Together, they form a $50-a-month habit that is easy to ignore and hard to justify to anyone who looks at a bank statement carefully.

The obvious move is to cancel something. The hard part is figuring out what. Dropping the chatbot is painful because it gets used daily. Dropping the cloud plan means either deleting data or migrating it, and migrating 340GB across ecosystems is exactly the kind of project that gets postponed indefinitely.

What was actually needed was not another subscription. It was a way to make the existing data smaller, more findable, and less intimidating.

Why Third-Party AI Cleanup Tools Felt Too Risky

The first instinct was to look outside the ecosystem. Surely some third-party AI service could connect to a Drive account, scan everything, and reorganize it with a few clicks. Several such tools exist, and some of them are genuinely clever.

Then reality intervened. A Drive account is not just a pile of memes and old screenshots. Mine contains a house deed, a copy of a will, business registration documents, and years of financial records. Handing a random startup broad, persistent access to that collection in exchange for tidier folders is a trade that becomes harder to justify the longer you think about it.

That left the possibility of a first-party tool built by the storage provider itself, operating inside the same trust boundary as the files. Which is exactly what Google quietly shipped.

Organize My Files Arrives, Quietly

Google's "Organize My Files" feature uses Gemini to analyze loose files in Drive and propose either moving them into existing folders or creating new folders for related clusters of files. Crucially, nothing moves automatically. Every suggestion is presented for review, and the user retains full control over what gets approved.

There are, however, meaningful gates. The feature is currently limited to Google Workspace accounts and Google AI subscribers, and Workspace smart features must be enabled for the button to appear in Drive at all. That means the people most likely to benefit from automated cleanup, namely free-tier users drowning in 15GB of accumulated clutter, are precisely the people who cannot access it.

How the Tool Works, Step by Step

1. Open Google Drive

Log into a Google account, open Drive in a browser, and navigate to "My Drive." The feature is browser-based for now and does not appear in the same form on mobile.

2. Click Suggest File Moves

Near the top of the file and folder list, a button labeled "Suggest File Moves" appears. Google has said it surfaces both in My Drive and inside parent folders, which theoretically allows users to drill into subfolders and get context-aware suggestions for each one.

Clicking it opens the Organize My Files panel, where Gemini begins analyzing loose files and generating cleanup recommendations. The analysis takes roughly a minute for a large account.

3. Review Gemini's Suggestions

Recommendations arrive in two flavors. The first is a proposed move of existing files into folders that already exist. The second is a proposal to create entirely new folders for groups of related files. Every file and folder in a suggestion can be previewed through a hovercard or opened in a new tab for a closer look before committing.

This review step is the most important part of the design. It keeps a human in the loop and prevents an AI hallucination from quietly relocating a legal document into a folder called "Misc 2024."

4. Approve the Changes

Checkboxes allow individual files and folders to be selected or deselected. Suggested folder names can be renamed if they are awkward or unclear, and destination folders can be changed if the AI guessed wrong. Once the selection looks right, approving the batch executes every move at once and returns the user to My Drive.

The Results: 19 Suggestions for 340GB

After the analysis finished, Gemini proposed 19 moves. Nineteen. Against a Drive containing years of clutter distributed across potentially thousands of folders, that is a rounding error.

Worse, the suggestions skewed heavily toward recency. Nearly everything Gemini surfaced was a file created or uploaded in the past several months, which is precisely the data least in need of rescue. The genuinely buried material, years-old drafts, orphaned downloads, duplicate exports, never came up.

Some suggestions were sensible. Gemini correctly identified a resume and a couple of resumes created for family members and proposed consolidating them into an existing resume folder. It suggested a new "Family and Real Estate" folder for property documents, and a "Travel Planning" folder for upcoming summer trip itineraries.

Then it grouped a file literally named "Delete" into the Travel Planning folder. That document was so named because it was marked for deletion. The AI did not recognize the intent, did not flag it as trash, and did not suggest removing it. It simply filed it alongside the itineraries.

Running It Twice Revealed a Bigger Problem

The changes were approved anyway, since even marginal improvement beats none. Then the tool was run again out of curiosity, on the freshly reorganized Drive.

Thirty seconds later, Gemini returned the exact same list: the same file moves, the same new folders, the same recommendations for files that had just been moved into those folders moments earlier. The tool appeared not to register that its own suggestions had already been executed.

That behavior points to a shallow scan rather than a persistent understanding of the Drive's structure. A genuinely useful assistant would learn from previous passes, remember which suggestions were accepted or rejected, and progressively work deeper into the backlog. Instead, it repeats itself.

What a Real AI Cleanup Assistant Would Need

The gap between what shipped and what users in this situation actually need is wide, and it is instructive to spell out.

• Age-aware prioritization. A cleanup tool should start with the oldest, least-touched files, not the newest. Recency bias makes the output feel cosmetic.
• Trash recognition. Files named "Delete," duplicates, expired drafts, and stale exports should be flagged for removal or archival, not filed neatly into folders.
• Cross-folder awareness. Asking users to run the tool in a thousand different parent folders and hope for fresh results each time is not a workflow. It is a chore.
• Memory between sessions. An assistant that forgets its own accepted suggestions within 30 seconds cannot build trust.
• Storage impact estimates. Users trying to shed terabytes care about how much space a cleanup actually reclaims. Number of moves is a poor proxy.
• Access for free-tier users. The people with the most painful storage problems are gated out of the tool designed to solve them.

An Industry-Wide Problem, Not Just One Product's

It is worth noting that this pattern is not unique to a single feature. Across the productivity software industry, AI assistants are being bolted onto file management, email, and note-taking with impressive demos and underwhelming depth. The models are capable of summarizing and classifying, but most implementations still lack the persistent memory and structural understanding required to manage a decade of accumulated personal data.

Cloud storage in particular is a hard problem because the value of a cleanup is inversely proportional to how easy it is. Someone with 20 files needs no help. Someone with 300,000 files needs so much help that a tool proposing 19 changes is effectively offering nothing.

There is also a real privacy architecture question underneath all of this. Users are being asked to let AI models crawl personal documents, legal records, and financial files in exchange for organizational convenience. That bargain only makes sense if the results are transformative. When the output is 19 suggestions for 340GB of data, the calculus flips quickly.

Where This Leaves the Subscription Math

Because the cleanup was so limited, the underlying problem remains untouched. The Drive is still messy, the storage total is still somewhere in the neighborhood of 340GB, and the monthly plan still renews at around $20 for 5TB of space that is mostly filled with files nobody can find.

The tool did just exit beta, and it is entirely possible that Google will deepen how Gemini scans Drive, prioritize older content, recognize obvious trash, and surface more aggressive organization suggestions over time. The architecture supports it. The current implementation simply does not deliver it yet.

What the experience makes clear is that automated file organization is one of the more genuinely useful applications for AI in productivity software, and one of the hardest to do well. It requires more than a language model that can read a filename and guess a category. It requires memory, hierarchy awareness, deletion judgment, and a willingness to touch the oldest and most neglected corners of an account rather than the newest and cleanest ones.

Until a tool does those things, the mess stays. So does the recurring charge.

Source: ZDNET News

Key Facts

• Only three standalone AI content detectors achieved perfect scores in the latest round: Pangram, QuillBot, and ZeroGPT.
• Several previously top-performing detectors declined, including Copyleaks, Originality.ai, and Undetectable.ai.
• Chatbots outperformed dedicated detectors: ChatGPT Plus, Copilot, and Gemini all scored perfectly on a five-text test.
• The ChatGPT free tier identified the writer of one human-written sample, raising privacy and identification concerns.
• Grok struggled, getting three of five classifications wrong and labeling all samples as human-written.
• False positives remain a serious problem, especially for non-native English writers.
• Plagiarism definitions apply to uncredited AI-generated text, even if no source was directly copied.

Why AI content detection is harder than it looks

The rise of generative AI since late 2022 has created a new kind of plagiarism problem. Merriam-Webster defines plagiarize as to steal and pass off the ideas or words of another as one's own, or to use another's production without crediting the source. That definition fits AI-generated content when a user presents machine-written text as their own without disclosure. The writer may not have copied a specific source, but the words were not theirs, and the true author, the AI model, goes uncredited.

Detecting that kind of text is not a simple technical task. AI detectors look for statistical patterns such as perplexity, burstiness, and token predictability. Human writing tends to be less uniform, with unexpected word choices and uneven sentence lengths. AI writing often is smoother and more predictable. But these signals overlap. A human editor can produce polished, formulaic prose. An AI model can be prompted to write with deliberate errors or varied rhythm. The result is an arms race between detectors and tools that claim to humanize AI text.

To test the current state of the art, a long-running independent evaluation used five blocks of text. Two were written by a human, and three were generated by ChatGPT. Each block was fed separately to each detector. A correct classification counted as a pass; an incorrect one counted as a failure. When a detector returned a percentage, anything above 70 percent was treated as a strong probability and taken as the detector's answer. The same five blocks were used across 11 standalone detectors, producing 55 individual tests.

The tested detectors included BrandWell, Copyleaks, GPT-2 Output Detector, GPTZero, Grammarly, Monica, Originality.ai, QuillBot, Undetectable.ai, Writer.com, ZeroGPT, and a newcomer called Pangram. Monica was dropped because it limited free tests to 250 words and then required a $200 upgrade to continue. Pangram was added and immediately became one of the top performers. The test series has now been run six times since early 2023, allowing some comparison over time.

Overall standalone detector results

The latest results were mixed. Only three detectors earned a perfect score: Pangram, QuillBot, and ZeroGPT. Several others landed at 80 percent: Copyleaks, GPTZero, and Originality.ai. GPT-2 Output Detector scored 60 percent. BrandWell, Grammarly, and Writer.com each scored 40 percent. Undetectable.ai had the weakest showing at 20 percent.

The declines were notable. In earlier rounds, as many as five detectors had perfect scores. In April 2025, Copyleaks, QuillBot, Undetectable.ai, ZeroGPT, and others were performing at or near the top. By this latest round, Copyleaks and Originality.ai had fallen. Both flagged human-written text as AI-generated. Copyleaks had publicly claimed 99 percent accuracy backed by third-party studies, yet it identified a human-written sample as 100 percent AI. Originality.ai, which sells usage credits and describes itself as a most accurate AI detector, also labeled the human sample as AI-written.

Other tools showed different problems. GPTZero, which has grown from a bare-bones project into a company with a mission of protecting what is human, swapped its errors from one round to the next. In April, it got the first human sample wrong and the second AI sample right. In the latest test, it got the first human sample right and the second AI sample wrong. That kind of instability is difficult for users to trust.

Grammarly, a well-known writing assistant, showed no improvement in AI detection. Its tool labeled an entirely ChatGPT-written sample as human. Writer.com did the same for every block, even though three of the five samples were machine-generated. Undetectable.ai, which also offers a service to humanize AI text, rated a human-written sample as 60 percent likely AI and rated three AI-written samples as 75 to 77 percent likely human. Those results were the opposite of accurate.

On the positive side, ZeroGPT has matured. Early versions looked sketchy, with no clear company name and lots of ads. Now it presents as a typical software service with pricing, contact information, and a professional interface. Its accuracy rose from 80 percent to 100 percent and held there. QuillBot, which once gave wildly inconsistent results across multiple passes, has now achieved perfect scores in consecutive rounds. Pangram, founded by former engineers from major tech companies, offers five free tests per day. Its processing was slow, with a partially white screen between scan and result, but it scored five out of five.

The test series also looked for a broader trend over time. So far, there is no strong pattern of consistent improvement. Test 5 was reliably identified as human across detectors and dates, but even that reliability declined in the latest run. The only safe conclusion is that detector performance remains volatile.

Chatbots as content detectors

The latest round added a new experiment: using general-purpose chatbots as content detectors. Each chatbot was given the same prompt, followed by the text to check: Evaluate the following and tell me if it was written by a human or an AI

Source: ZDNET News

The outcome was unusually decisive. Google’s Nano Banana Pro, part of Gemini 3, earned a near-perfect 93%. ChatGPT placed second at 74%. Four other tools — Midjourney, Adobe Firefly, Leonardo AI, and Canva — finished between 43% and 54%. The spread suggests that the best image generators are not merely ahead by a small margin; they are operating at a different level.

Key facts at a glance

• Winner: Google Nano Banana Pro, part of Gemini 3, with a 93% score.
• Runner-up: ChatGPT image generation, 74%.
• Other scores: Midjourney 57%, Adobe Firefly 54%, Leonardo AI 52%, Canva 43%.
• Cost of winner: $19.99 per month as part of Google AI Pro.
• Test categories: revising existing images, generating original images, adding text, and leveraging pop culture.
• All six tools performed well when asked to create a simple logo. Beyond that, performance diverged sharply.
• Common weaknesses: mangled text, distorted faces, poor photo cleanup, and inconsistent pop-culture handling.

How the test worked

The evaluation used the same nine prompts across all six generators. The prompts included: dressing a man in a US Navy admiral’s uniform on an aircraft carrier bridge; restoring a black-and-white photo; restoring and colorizing a black-and-white truck photo; creating a retro-futuristic logo for a video studio named Space Coast Studios; generating a medieval librarian in a candlelit stone library; creating a photorealistic senior adult holding a flagship smartphone with Facebook-style ad copy; creating a candid student portrait with a MacBook Pro and coffee; designing a poster for a fictional fourth Back to the Future movie set in 1920s New York; and generating a Nightmare Before Christmas-style IT professional in a data center.

Each image was scored on specific criteria. For recontextualization, points depended on proper background, appropriate new clothing, and whether the original subject’s face and clothing were preserved. For restoration, points were given for keeping original elements, cleaning up damage, preserving facial features, and handling colorization. For logos, text accuracy and thematic fit mattered. For social posts, text placement, legibility, and realistic anatomy were considered. For pop culture, recognizability, style, and textual accuracy were judged.

Google Nano Banana Pro: the clear winner

Best AI image generator overall by a wide margin

• Overall score: 93%
• Cost: $19.99 per month as part of Google AI Pro

Nano Banana Pro dominated nearly every category. In the admiral recontextualization test, it kept the subject’s face and glasses, changed the angle he was facing, placed him properly on a ship bridge, and added binoculars to his hands. That level of identity preservation and scene coherence is rare among image generators.

It also performed well on old photo restoration. Given a dark childhood photo, it produced a cleaner, more visible image while preserving the original subject. When asked to restore and colorize a truck photo, it mostly preserved the lettering, though it introduced a small error: the text read something close to “BADICLOGICAL DEFEKSE” instead of “RADIOLOGICAL DEFENSE.” Even so, the result was far more accurate than most competitors.

Creative generation was another strong point. For the Space Coast Studios logo, Nano Banana Pro included a palm tree, a rocket, film imagery, and readable text. For the medieval librarian, it produced candlelight, stone walls, and a convincing character. For social media, it generated a senior adult holding an iPhone-like flagship smartphone with clean, correctly spelled text. It also created a student with a MacBook Pro and coffee in a cozy setting. Interestingly, when asked for a flagship smartphone, Google’s own AI generated what was clearly an iPhone.

Pop-culture prompts revealed the biggest gap. Nano Banana Pro produced a fictional Back to the Future poster with a DeLorean, a skateboard, period clothing, and a marquee reading “New York City 1925.” It added the tagline “The Roaring Twenties Got Heavy,” evoking both the era and a character’s favorite phrase. Its Nightmare Before Christmas-style IT professional also captured the Tim Burton aesthetic. The main drawbacks were a watermark in the bottom corner and one session where it confused an earlier subject with a truck cleanup request. It also occasionally cuts off free users after a limited number of images. Still, with a 93% score, it was the only tool that came close to acing the entire suite.

ChatGPT: strong chat iteration, inconsistent image results

Best for natural-language iteration in chat

• Overall score: 74%
• Cost: $20 per month as part of ChatGPT Plus

ChatGPT’s image generator, now part of the ChatGPT Images update, ranked second. Its biggest advantage is conversational iteration: users can refine an image through natural language in a chat interface. That flexibility makes it useful for brainstorming and quick edits. However, its image quality was uneven across the test suite.

Photo retouching was not a strength. The admiral recontextualization looked like the original subject, but the uniform insignia were unclear and seemed to mix ranks. The childhood photo restoration was good, preserving the original face rather than inserting an adult’s face onto a child’s body. The truck cleanup was acceptable but changed “RADIOLOGICAL DEFENSE” to “RADIO CHEMICAL DEFENSE,” with odd spacing.

Logo generation was fine, though it lacked Florida-specific cues like palm trees or sun. The librarian image met expectations. Social media posts were mostly correct, but both text blocks were placed in the same position, and the light font and poor contrast reduced readability. Points were deducted.

Pop-culture handling was the most inconsistent. A Star Trek-inspired prompt produced excellent results, even duplicating a creature’s man-in-suit look. But Back to the Future and Nightmare Before Christmas prompts were refused or mishandled. That inconsistency dropped the new version from 81% to 74%. The previous version had produced a decent Back to the Future image with a recognizable DeLorean and period city, though the Nightmare IT character had a questionable third hand. For now, ChatGPT remains a solid all-around assistant but not the most reliable dedicated image generator.

Midjourney: cinematic creativity, limited cleanup

Best for hyper-creative cinematic imagery

• Overall score: 57%
• Cost: $10 per month for Midjourney Basic

Midjourney has long been a favorite for artists and designers. It excels at original creative imagery, mood, and cinematic style. In this test, it produced a librarian scene that was almost exactly what was requested. Its social media portraits looked realistic, with believable skin and hands. Its Nightmare Before Christmas-style IT image was a standout: a photo-realistic Lurch-like character in a data center, mixing gothic horror with corporate infrastructure.

Where Midjourney struggled was in editing existing images. The admiral recontextualization did not preserve the subject’s identity. The childhood photo was not cleaned up and was arguably made worse. The truck prompt produced a different truck altogether, and inexplicably included a woman resembling a Star Trek captain. Midjourney also failed to render text on the social media posts, losing points. Its Back to the Future poster had no text and looked too neon for 1920s New York. For pure artistic generation, Midjourney remains powerful. For practical photo editing and text, it lags far behind.

Adobe Firefly: commercial safety at a cost

Best for commercial-safe images

• Overall score: 54%
• Cost: $9.99 per month as part of Firefly Standard, also included in many Adobe plans

Adobe Firefly is designed with copyright and commercial safety in mind. It powers AI features inside Photoshop and other Adobe tools, and it also works as a standalone web interface. Users can now choose other engines, including Nano Banana, but this test used Firefly Image 5 preview.

Firefly was the most restrictive generator. It refused to upload a personal photo because the subject wore a Star Trek-inspired Gorn T-shirt. The image had to be altered by another AI before Firefly would accept it. That caution may appeal to commercial users, but it creates friction. The admiral image barely resembled a uniform and did not preserve the face. The childhood photo transformation was good and visible. The truck colorization looked great, but Firefly did not clean up image artifacts, which was the main task. Logo generation was usable. The librarian image suffered from an uncanny valley face. Social media prompts triggered an error about artist names because the phrase “Facebook-style” was flagged. After removing it, Firefly generated images, but one had a spelling error and the student image had subtle anatomical problems. Pop-culture prompts were simply refused.

Leonardo AI: fantasy strength, editing refusal

Best for fantasy art

• Overall score: 52%
• Cost: $10 per month, also part of Canva Business plan

Leonardo AI refused the existing-image tests entirely. It would not attempt the admiral, restoration, or colorization tasks. That immediately limited its score. It performed better with original creation. The logo was workable. The librarian met the open-ended requirements. Social media posts had issues: the man’s inner elbow looked odd, and the woman’s leg wrapped around the laptop while she balanced coffee on her leg.

Leonardo’s Back to the Future prompt was notable because it was the first tool to actually format a poster. However, it invented sequels four through ten, produced gibberish text like “HAMY JOB OARY OF DOSE,” and included cars and buildings that were not from the 1920s. Its Nightmare Before Christmas-style IT image was a favorite: it captured the gothic feel and data center setting with a photo-realistic Lurch-like character. For fantasy and stylized art, Leonardo has clear strengths, but it is not a general-purpose image editor.

Canva: marketing integration, weak generation

Best to generate art right inside marketing designs

• Overall score: 43%
• Cost: $15 per month for Pro or $20 per month for Business, which includes Leonardo

Canva is a dominant force in marketing design. Its acquisition of Affinity Pro and release of free Affinity tools expanded its creative suite. Canva also offers its own generative AI features, separate from Leonardo, though Canva owns both. The Business plan includes Leonardo, making it an attractive bundle for marketers.

Canva’s image generator finished last. Its photo cleanup and recontextualization results were bizarre. The admiral image appeared to have trees growing on an aircraft carrier. It identified a beard but failed to include the requested truck. A disembodied woman’s head floated above wreckage that looked like a cross between a submarine and a Tesla tower. Logo and librarian images were fine, and logo generation was the only category where Canva earned full points. Social media posts were disappointing: text spacing was off, and the woman’s image had text that lost its way completely. Once again, coffee was balanced on a leg.

For pop culture, Canva produced a Back to the Future 4 poster with a Marty-like character, a DeLorean, and the proper logo format. But the text was gibberish, a girl appeared, an object floated in the sky, generator-like devices lined the road, and half the road was on fire. It also denied a Nightmare Before Christmas-style prompt even though it had allowed the Back to the Future reference. That inconsistency cost points and highlighted the uneven guardrails in its system. Canva’s inconsistent handling of a Back to the Future poster versus a Nightmare Before Christmas prompt remains one of the clearest examples of why the scoring gap is so wide.

Source: ZDNET News

Closing the books as a small-business owner should not feel like a high-stakes survival game. Yet many entrepreneurs still struggle with spreadsheets, shoebox receipts, and a pile of invoices that never seems to shrink. Before long, paperwork can become overwhelming right before tax season.

Modern accounting software has moved far beyond basic bookkeeping. This guide breaks down the best small-business accounting platforms of 2026 after months of testing. The picks are based on real-world scenarios, automation quality, mobile access, integration depth, support resources, and value for money.

Key Facts at a Glance

• Best overall: QuickBooks Online for its powerful automation, broad integration ecosystem, and scalable feature set.
• Best for freelancers and service-based businesses: FreshBooks for time tracking, project management, and client-friendly invoicing.
• Best for growing businesses: Xero for modern design, unlimited users, and a large app marketplace.
• Best for AI-powered startup accounting: Puzzle.io for automated transaction categorization and startup metrics.
• Best for enterprise-grade financial management: Sage 50 Accounting for desktop control, advanced reporting, and industry-specific modules.
• Best free option: Wave for basic invoicing, expense tracking, and reporting at no cost.
• Best all-in-one suite option: Zoho Books for tight integration with other Zoho business apps.

What Is the Best Small-Business Accounting Tool Right Now?

QuickBooks Online is the top pick for most small businesses in 2026. It is widely recommended because it balances powerful features with a user-friendly design. It includes inventory management, mobile apps, and integration with more than 750 third-party tools through its app store. Plans scale with the business, whether the owner is tracking basic expenses or managing complex multilocation operations.

The evaluation focused on five critical factors: learning curve for non-accounting business leaders, mobile accessibility, integration with other business software, features that can replace separate tools, and overall value for money. Platforms that offer strong support and learning resources were prioritized because many small-business owners are learning accounting as they go.

The Best Small-Business Accounting Software in 2026

QuickBooks Online: Best Overall

QuickBooks Online dominates the small-business accounting space for good reason. It simply works for most scenarios. Its bank-feed automation is impressive, automatically categorizing transactions and flagging duplicates with high accuracy. The platform’s AI assistance saves time by suggesting expense categories and providing smart guidance throughout the workflow.

There is also an app store with more than 750 integrations, so it plays nicely with virtually any business tool already in use. The mobile app deserves special mention: users can capture receipts, send invoices, and check cash flow from anywhere. There is a learning curve, but the payoff comes quickly once the interface becomes familiar.

QuickBooks Online’s pricing reflects its comprehensive nature. It starts higher than some competitors but delivers enterprise-level features to small businesses. For growing companies that need solid reporting, multiuser access, and accountant collaboration, QuickBooks Online justifies its premium positioning. It is particularly strong for businesses that need inventory tracking and have complex workflows.

Pros: Lots of integrations; powerful automation capabilities; industry-standard choice.

Cons: Initial learning curve; overwhelming feature set; expensive subscription pricing.

Key features: Bank reconciliation, AI-powered categorization, multicurrency support, mobile receipt capture, payroll integration, custom invoicing, and real-time reporting.

FreshBooks: Best for Freelancers and Service-Based Businesses

FreshBooks excels at what matters most to service providers: turning time into money. It seamlessly converts tracked hours into professional invoices, eliminating the tedious manual process most freelancers endure. Its client portal feature creates transparency that strengthens relationships while reducing back-and-forth emails about project status.

This platform is best suited for creative workers, consultants, and small agencies with hourly billing. Its proposal-writing tools outperform many dedicated proposal platforms, letting users create project pitches right inside the accounting software. Automation works reliably when configured properly, though occasional glitches still require manual intervention.

Where FreshBooks stumbles is scalability. Adding users costs about $11 per person monthly, making the software expensive for growing teams. Inventory management remains basic compared with QuickBooks, limiting its appeal for product-based businesses. Customer support has declined recently, with slower response times for most queries.

Pros: Intuitive interface; time tracking tools; project management built in.

Cons: Limited scalability; basic inventory; higher user costs.

Key features: Built-in time tracking, project collaboration, automated invoicing, expense management, client portal, proposal creation, and mobile apps.

Xero: Best for Growing Businesses

Xero represents the modern evolution of cloud accounting, with a design that makes financial management more enjoyable. Its bank reconciliation process is consistently impressive: connecting accounts takes minutes, and automatic transaction matching works flawlessly. There are also no per-seat pricing headaches, because unlimited users can be added to all plans.

The platform’s strength lies in its huge app marketplace, with more than 1,000 integrations that let businesses build customized accounting ecosystems. Xero’s multicurrency capabilities handle international transactions well, automatically updating exchange rates and managing foreign invoices. The mobile app maintains full functionality, letting users approve bills, send invoices, and check cash flow while traveling.

However, Xero’s pricing sits on the higher end, especially for businesses with fewer users. The lack of phone support frustrates users who prefer immediate assistance over email-only channels. Although payroll integration exists through third-party providers, native payroll features remain limited compared with QuickBooks.

Pros: Modern interface design; largest integration marketplace; unlimited users.

Cons: No phone support; limited payroll features; higher pricing.

Key features: Unlimited users, 1,000+ app integrations, multicurrency support, project tracking, bank reconciliation, mobile accessibility, and real-time reporting.

Puzzle.io: Best for AI-Powered Startup Accounting

Puzzle.io is a next-generation accounting tool built specifically for startups. Its AI-powered transaction categorization achieves 90% to 95% accuracy by learning from business patterns. Built-in tools automatically draft reconciliations and financial statements, freeing founders to focus on growth.

Beyond traditional accounting, Puzzle.io can track burn rate, runway, annual recurring revenue, and monthly recurring revenue, the metrics that matter most for venture-backed companies. The real-time dashboard provides daily financial updates instead of stale monthly reports, giving users the agility needed in fast-moving startup environments. It handles both cash and accrual accounting at the same time, eliminating manual spreadsheet work.

The integration ecosystem connects with modern tools such as Stripe, Brex, Ramp, and Gusto, creating a unified financial data source. However, Puzzle.io’s laser focus on startups means it is not ideal for all types of small businesses. It is also relatively new compared with established players, which might concern those seeking stability.

Pros: AI automation; real-time insights; startup focused.

Cons: Limited customization; newer platform; startup-centric only.

Key features: AI transaction categorization, real-time financial dashboards, accrual automation, startup metrics tracking, bank reconciliation, tax compliance, and partner integrations.

Sage 50 Accounting: Best for Enterprise-Grade Financial Management

Sage 50 Accounting is a desktop-first solution with cloud connectivity, making it ideal for local data control with online accessibility. It can generate more than 150 different reports instantly, from basic financial statements to complex inventory analyses. Its reporting is among the most comprehensive available in accounting software.

But Sage 50’s complexity creates a steep learning curve for new users. Its pricing reflects the comprehensive feature set, starting higher than any small-business-focused competitor. Users can handle complex manufacturing workflows and even replace payroll software entirely, but the trade-off is a less modern user experience. Customer support quality varies, with some users reporting difficulty getting timely technical assistance.

Sage 50 works best for established businesses that have complex accounting needs and value features over simplicity. Its industry-specific versions for construction, manufacturing, and distribution provide specialized tools that generic accounting software lacks. Most small businesses will find these features to be overkill for their requirements.

Pros: Solid reporting; local data control; industry-specific modules.

Cons: Much higher costs; dated interface; limited cloud features.

Key features: Advanced inventory management, job costing, multicurrency support, comprehensive reporting, payroll integration, bank reconciliation, and industry-specific tools.

Wave: Best Free Accounting Option

Wave is a cloud-based accounting tool for freelancers and small businesses. It offers basic invoicing, expense tracking, and financial reporting completely free of cost. It is a practical starting point for solo operators who need simple bookkeeping without a monthly subscription, though it lacks the deeper automation and support of paid platforms.

Zoho Books: Best for All-in-One Suite Users

Zoho Books is part of the Zoho business suite. It integrates easily with other Zoho apps, making it a strong choice for startups and small teams looking for an all-in-one package. Businesses already using Zoho CRM, Zoho People, or Zoho Expense can centralize workflows and reduce the need for separate subscriptions.

How the Picks Compare

• QuickBooks Online: Starts at $19 per month; extensive custom fields and workflows; 750+ integrations; intuitive interface.
• FreshBooks: Starts at $21 per month; limited customization; moderate integration support; very user-friendly.
• Xero: Starts at $29 per month; good customization; 1,000+ app integrations; modern and intuitive.
• Puzzle.io: Starts at $25 per month; limited, startup-focused customization; modern tool integrations; AI-powered simplicity.
• Sage 50 Accounting: Starts at $62 per month; extensive industry-specific options; limited integrations compared with cloud platforms; requires training and a complex interface.

Which Accounting Tool Is Right for You?

• Choose QuickBooks Online for comprehensive features, extensive automation, and the largest ecosystem of integrations for growing businesses.
• Choose FreshBooks for simple time tracking and project management focused on freelancers and service-based businesses.
• Choose Xero for modern cloud accounting with unlimited users and strong international capabilities for collaborative teams.
• Choose Puzzle.io for AI-powered automation specifically designed for startups tracking metrics like burn rate and runway.
• Choose Sage 50 Accounting for desktop-based comprehensive business management with advanced inventory and manufacturing capabilities.

Factors to Consider When Choosing Small-Business Accounting Software

Do not blindly choose whichever vendor seems best right now. Balance immediate needs with future growth potential. Several factors matter when selecting an accounting tool to support a business through its early stages.

• Growth accommodation: The software should handle increasing transaction volumes, additional users, and more complex workflows. Look for tiered pricing that lets you upgrade features without migrating to a different system.
• Integration ecosystem: Most modern businesses use multiple types of software, so the accounting tool must connect with payment processors, banks, CRM systems, ERP applications, and more. Look for platforms with more than 500 prebuilt integrations, plus custom development APIs.
• Data protection, privacy, and compliance: All cloud-based software should include two-factor authentication, data encryption in transit and at rest, and GDPR compliance features as standard. Also check for data breaches in the software’s history.
• User experience design: For a small to midsize business with minimal staff, the accounting tool should be easy enough for non-accountants to use. Consider how much training the team will need and whether there are enough onboarding resources.
• Customer support quality and availability: Look for vendors offering multiple support channels, including phone, chat, and email, with reasonable response times. Check reviews that mention support quality. A slow resolution could delay payroll or freeze a quarterly close.
• Total cost of ownership: Subscription costs never give the full picture. Factor in setup costs, user additions, integration fees, and potential upgrade expenses. Some platforms charge extra for 1099 processing, which should also go into calculations.
• Industry-specific features: Businesses in construction, retail, or professional services need specialized tools like job costing, inventory management, or project tracking. Make sure the accounting software meets the industry’s tax and reporting standards.

How the Testing Was Done

Each platform was tested across real-world scenarios that small-business owners encounter daily. Sample companies were set up, common transactions were processed, and the time required to complete tasks was tracked. Automation was studied closely: how well each platform categorized expenses, reconciled bank feeds, and generated reports without input.

Practical usability was prioritized over flashy feature lists. Mobile apps were tested extensively because business owners need access to financial data on the go. Customer support was another key factor, since quick resolution of technical issues is vital for growing operations.

FAQs on Small-Business Accounting Software

What features should I look for in my accounting software?

It depends on your industry, growth stage, and business setup. Standard features include invoicing, expense tracking, bank reconciliation, and financial reporting. Advanced platforms also offer inventory management, payroll integration, and multicurrency support, which can save costs on other tools.

Can accounting software help with tax preparation and compliance?

Yes, accounting tools can simplify tax preparation by automatically organizing income and expenses into proper categories. Most platforms generate all the reports an accountant needs, with many even including built-in tax compliance features.

Do you need accounting knowledge to use small-business accounting software?

Modern accounting tools are designed to be user-friendly for non-accountants, using simple terminology and step-by-step guidance. That said, users still need to understand some basic terminology and core accounting concepts. Things get considerably more complicated as a business moves upmarket.

Should I choose cloud-based or desktop accounting software?

Cloud-based tools offer automatic updates, mobile access, and easier collaboration. On-premise desktop software provides more control over data. The choice often comes down to balancing security and ease of use.

I already have a bookkeeper. Is it still worth it to get accounting software?

Absolutely. Most accounting software lets users grant access to a bookkeeper so they can work directly with financial data. Many platforms also include special accountant modes with additional features that help professionals manage books more efficiently. This can speed up tax preparation and financial reporting with fewer chances of errors.

For businesses comparing alternatives, Wave offers a free entry point for basic needs, while Zoho Books provides deep integration with the wider Zoho suite, including CRM, payroll, and expense tools, which can reduce the need for separate subscriptions.

Source: ZDNET News

The integration is taking place within the manufacturing environment rather than in consumer products. Samsung is using Mistral AI large language models to monitor production data across wafer fabrication facilities. Those facilities generate data from thousands of sensors, equipment sensors, metrology tools, and process control system logs. Each silicon wafer passes through hundreds of stages, and any slight anomaly can affect yield, performance, or reliability.

Mistral AI models are being applied to identify anomalous patterns in that data and to assist engineers in diagnosing root causes. The key benefit is speed. On a modern production line, the time between detecting a defect and adjusting process conditions can be measured in hours. AI-assisted tools can compress that window by highlighting the most relevant variables for engineers to examine. The approach also supports preventive action, catching drift in equipment performance before it reaches the threshold that results in scrap.

• Samsung has adopted Mistral AI models in semiconductor manufacturing operations.
• The AI supports defect detection, yield improvement, and equipment control.
• Mistral AI is a French artificial intelligence startup known for powerful and efficient large language models.
• The decision underscores a broader industry shift toward AI-driven manufacturing.

Why AI is becoming critical in chip fabrication

Semiconductor manufacturing is one of the most complex production processes ever developed. Advanced chips are built on wafers through processes involving lithography, deposition, etching, cleaning, and polishing. At leading-edge nodes of five nanometers and below, even subatomic-level variations can lower the number of functional chips produced by each wafer. Because the revenue in semiconductor manufacturing depends almost entirely on yield, chipmakers must continuously refine their processes.

Traditionally, engineers relied on statistical process control and rule-based systems to set acceptable limits for equipment conditions. That approach still works but is limited when interactions between multiple variables become too complex for a human team to predict. A modern fab may have tens of thousands of process parameters. Some effects only appear after multiple processing stages. AI models can identify correlations that are not obvious from raw data or conventional charts.

Machine learning has been used in semiconductor manufacturing for years, but the rise of foundation models changes what is possible. Large language models can combine numeric data with textual logs, maintenance records, and equipment manuals. Instead of simply flagging an out-of-range sensor reading, an AI assistant can generate a longer narrative around a problem, describing likely causes, suggesting tests, and summarizing similar past incidents.

That kind of reasoning capability is attractive to Samsung. The company has made no secret of its ambition to improve the efficiency of its fabs. Like every chipmaker, Samsung runs its plants around the clock. Any improvement in predictive maintenance or process matching has an immediate financial impact. By adding Mistral AI's models to its toolbox, Samsung is effectively deploying a flexible reasoning layer across operational data sets.

Samsung's foundry push and competitive pressure

Samsung's adoption of Mistral AI models should be viewed in the context of its broader semiconductor business. The company is one of the world's largest memory makers and a major player in contract chip manufacturing. In the foundry market, Samsung has fought to close the gap with TSMC, which controls a dominant share of advanced chip production. Samsung's ability to win orders from high-profile customers depends not only on process technology but also on consistent production results.

AI-powered yield improvements are therefore a strategic matter. The semiconductor industry has entered an era in which process node advantage alone is no longer enough. Customers evaluate a foundry on production quality, power efficiency, and time to market. A slight improvement in yield through AI analytics can strengthen Samsung's position in negotiations with fabless clients and data center vendors.

The same technology also applies to memory production. Samsung's DRAM and NAND flash factories are capital-intensive environments that generate terabytes of data every day. Smarter analysis can improve inventory forecasting, equipment scheduling, and quality control. Mistral AI is not tasked with replacing existing process control systems but with adding a layer of intelligence that can make sense of all the information flowing through a smart factory.

A European AI champion enters industrial manufacturing

Mistral AI was founded in 2023 by former researchers from Google's DeepMind and Meta's AI laboratory. The company quickly became a prominent European alternative to US technology giants. Its large language models are designed to deliver high performance with fewer computing resources than those of some rivals. That efficiency is particularly valuable when models are deployed inside an industrial facility, where hardware budgets and energy constraints are real.

Unlike some top-tier models that are only available through cloud APIs, Mistral AI has released open-weight models. Those models can be downloaded and run on private servers. For a chipmaker like Samsung, that flexibility is important. Semiconductor manufacturing data is proprietary and extremely valuable. Sending such data to an external AI service may be impossible for security and confidentiality reasons. Running models within Samsung's own infrastructure allows engineers to take advantage of artificial intelligence without exposing sensitive process recipes.

Use of a European AI startup also gives Samsung access to a development community that is actively pushing open model innovation. Mistral AI has demonstrated strong performance on reasoning tasks and code generation. Industrial engineering teams often need models that can understand technical language and interpret charts. The capability to process both natural language and structured data makes Mistral's models a practical fit for factory environments.

What this means for the semiconductor industry

The semiconductor sector is being watched closely as a test case for physical-world AI applications. Robots have been used in fabs for decades, and artificial intelligence has supported certain chores like defect classification. But the move by Samsung represents a deeper deployment of generative AI into operational workflows. That could open a path for many other manufacturers to follow.

The high cost of semiconductor design and manufacturing is pushing companies to look for every possible source of efficiency. Global chip sales are cyclical, but advanced manufacturing costs continue to rise. If generative AI can help shave even a few percentage points off defect rates or increase equipment utilization by one or two percent, returns can be enormous. The use of AI in fabs is not a futuristic idea. It is becoming a prerequisite for remaining competitive.

Samsung's work with Mistral AI may prove to be a small step, but the direction is clear. Chipmakers will continue looking for specialized AI models that can be securely deployed in local operations. Companies that control their own data and run AI where those chips are produced may have a major advantage in the next phase of the industry. As more fabs become fully digitized, the interaction between physical processes and intelligent software will only become more important.

Source: AI News News

Key facts at a glance

• Arm has extended its Total Design program to cover Physical AI and robotics.
• The framework is intended to accelerate custom silicon development for machines that interact with the physical world.
• It builds on the existing Neoverse-based Total Design ecosystem uniting foundries, IP vendors, EDA specialists and software companies.
• The program focuses on heterogeneous compute, combining CPUs with accelerators for perception, planning and control.
• Energy efficiency and software compatibility are central goals for battery-powered and thermally constrained robots.

What is Physical AI?

Physical AI is commonly used to describe artificial intelligence systems that are embodied in hardware and interact with the physical world. Unlike large language models that operate in the digital domain, physical AI must be grounded in sensors and actuators. It covers applications such as industrial robots, collaborative robots, autonomous mobile robots, drones and self-driving vehicles. These systems require real-time perception, planning and control, often at low energies, and need to respond to unpredictable environments with high reliability.

The demand for such systems is rising across industries. Logistics operators are employing autonomous mobile robots to move goods. Factories are using articulated robots embedded in increasingly automated production lines. Agriculture, construction and healthcare are testing robotic systems to handle repetitive, hazardous or precision-intense tasks. At the same time, the AI models that power those systems are becoming larger and more complex, which places new pressure on underlying compute architectures.

Extending the Total Design concept

Total Design was originally launched as an ecosystem program to support custom silicon development around Arm's Neoverse compute subsystem. It brought together leading foundries, IP providers, EDA vendors and cloud service providers to reduce the time and cost required to build high-performance, energy-efficient chips for data centre applications. The new Physical AI and robotics framework extends this model to the edge, where machines must process sensor data with low latency, adapt to changing environments and operate on battery power or otherwise constrained thermal budgets.

By extending Total Design to physical AI and robotics, Arm is targeting a market that has historically relied on off-the-shelf components. While many robotics prototypes use general-purpose processor boards, production systems often demand better tuning for power, size, weight and cost. Custom silicon can deliver improvements in all these areas, but only if designers have access to a wide set of building blocks. Arm can supply some of those blocks in the form of CPU core designs, interconnect IP and system-level architectures.

A fragmented market needs flexible compute

The company's ecosystem approach also acknowledges the increasing fragmentation of the robotics industry. Robots are used in vastly different ways, and no single chip design can serve every use case. A home companion robot has different requirements from a heavy-duty warehouse manipulator, and each of those in turn differs from a medical delivery robot. Rather than forcing hardware makers into a single architecture, Arm seeks to provide a flexible foundation that allows designers to construct tailored solutions while maintaining software compatibility across generations.

Software compatibility is a key part of this endeavor. Arm's instruction set architecture is already deployed in a wide range of devices, from tiny sensors to high-performance servers. That breadth gives robotic developers the opportunity to reuse software stacks across their product lines. The new framework aims to build on that advantage by aligning Arm's ecosystem partners around common software interfaces, performance benchmarks and security requirements.

Energy efficiency and autonomy

Energy efficiency also lies at the heart of Arm's strategy. Many robots are mobile and battery-powered. Even those that are plugged into an industrial power supply may need to satisfy strict heat and space constraints. Arm has long competed on power efficiency, and the company argues that this becomes even more important when autonomous machines are expected to operate for long periods of time without interruption. Efficient compute means more uptime, fewer charging stops and lower overall ownership costs.

Efficient compute also supports the growing trend toward on-device AI in robotics. Instead of sending every image or sensor reading to a cloud server, a robot can run its perception models locally, reducing latency and preserving privacy. The addition of specialized AI accelerators in a custom SoC can make this process dramatically more efficient. Arm's Total Design framework gives chip designers a clear pathway for integrating those accelerators with Arm CPUs, including the necessary software stacks and validation tools.

Industry implications

Arm's push into robotics is not entirely new. The company has long supplied processor IP for automotive systems and industrial microcontrollers. Many robotic controllers already use Cortex-A, Cortex-R and Cortex-M processors. However, the Total Design for Physical AI announcement appears to signal a more concerted effort to support high-performance, large-scale robotic systems that require a level of compute capability typically associated with servers. It is an attempt to capture a much bigger share of the value in emerging robotics platforms.

An important element of the expanded Total Design focus is likely to be a set of reference architectures. Rather than starting from a blank page, chip developers can choose from a range of validated Arm Neoverse-based configurations and customize them for a particular robot product. These reference designs can dramatically cut time to tapeout, enabling smaller companies to bring their own silicon to market without assembling dozens of expert teams. Larger semiconductor vendors may also use the framework as a starting point for broader portfolio exploration.

Underpinning the initiative is the notion that AI is moving from the data center to the edge and, ultimately, into machines that touch the physical world. Arm has been a platform provider for many shifting computing epochs, from mobile phones to cloud servers. The Total Design for Physical AI framework appears to be a bid to ensure it remains a central participant in the next wave of computing. The company's technology is already present in a vast range of devices, but winning the robot era will require a robust coalition of silicon, software and systems suppliers.

It remains to be seen how quickly custom robotic chips will proliferate. Many robotics manufacturers still prefer commercial off-the-shelf boards due to development cost and lead time. Yet as production volumes grow and workloads stabilize, the economics of custom silicon become more attractive. Arm appears ready to position itself as the connective tissue that lets large and small developers tap the advantages of custom compute without bearing an unrealistic burden.

With the new framework, the company is ultimately betting that the ecosystem approach that has served mobile computing so well can also serve physical AI. By lowering barriers for specialized chip development, Arm hopes to inspire a generation of robotic systems built on its architecture. The success of this endeavor will depend not only on technology, but on the ability of Arm's many partners to move together quickly toward a common goal. The robotics market has been waiting for an infrastructure moment like this, and Arm's framework could signal the start of a more mature and scalable era for physical AI development.

Source: AI News News

Malaysia's retail ecosystem includes modern hypermarkets, convenience stores and thousands of traditional provision stores. Many of these smaller outlets, in urban and rural areas alike, still operate with informal ordering processes. The lack of visibility into demand often leads to two opposite problems. A busy store may run out of a popular product because the owner underestimated demand. Another store may have plenty of stock but many units are sold close to their expiry dates, which affects the consumer experience and increases waste.

For Coca-Cola, the business case is clear. A small format retail outlet may only sell a few categories of beverages. If a consumer arrives and the desired brand is not on the shelf, they will often choose a competing drink. Out-of-stocks can therefore translate into direct loss of sales for Coca-Cola and a diminished reputation for the shop. By using AI to anticipate demand, both supplier and retailer can avoid that risk. At the same time, the retailer reduces tied-up capital in slow-moving inventory, and the supplier obtains a more stable and predictable order pattern.

What the AI ordering system does

The AI-based ordering tool is designed to function as a digital assistant for retailers. Rather than merely recording transactions, it learns from each outlet's behavior and uses that knowledge to make accurate replenishment suggestions. The system uses historical sales data, past order quantities, promotional calendars, weather patterns, holidays and even local community events to estimate likely demand in the days ahead.

For example, a shop located near a school might normally sell more small-size beverages and snacks. During school holidays, however, that same shop may see a sharp drop in lunchtime traffic. The AI model recognizes this shift and adjusts the recommended order accordingly. In another area, demand for larger PET bottles could increase before the weekend. The system weights all those variables and produces a single set of suggested delivery quantities for the retailer.

The retailer does not need to understand how machine learning works in order to benefit from it. The application presents the recommendation clearly and allows the user to change a quantity with a simple plus or minus button. A retailer who knows that an upcoming celebration will bring many visitors can increase the order of higher-margin or premium products. A retailer with limited storage can reduce the quantities of large packs if they do not fit in the store.

Once the retailer confirms the order, the information immediately travels to Coca-Cola's supply chain systems. From there, orders are aggregated by route, warehouse and delivery date. Trucks can be loaded in the most efficient sequence, and drivers can plan their routes based on actual customer demand. This has the potential to lower fuel use, reduce traffic time and allow more outlets to be served each day.

The importance of traditional trade in Malaysia

The traditional trade channel, also known as general trade, is still a significant part of beverage sales in Malaysia. Independent grocery stores, convenience shops and family-run stores serve communities in suburbs, small towns and the kampungs that dot the country. These businesses often lack the sophisticated inventory management systems used by large supermarket chains, but their locations and close relationships with customers make them extremely valuable to brands.

In many ways, the AI ordering system is more meaningful for these small shops than for large modern retail chains. Big supermarkets usually have procurement teams and software that can generate orders automatically. Small retailers do not have that resource. Their orders are often based on the salesperson's advice, the amount of cash available at the moment, or the amount of empty shelf space behind the counter. Coca-Cola's new tool puts the power of predictive analytics directly into the hands of these merchants.

The transformation is also important because Malaysian consumer tastes are not uniform. Demand for carbonated soft drinks, bottled water, ready-to-drink tea, coffee and fruit juice varies between states and even between neighborhoods within the same city. A model trained on data from one outlet can learn local preferences better than a national ordering manual. This means the system can propose the right assortment for every store. A small store in a predominantly residential area may receive a recommendation for more family-size drinks, while an outlet near an office district may see a higher number of single-serve bottles.

How the AI model improves over time

One of the most valuable characteristics of AI is its ability to improve through experience. Each time the retailer accepts the recommended order, the model treats that as confirmation that its forecast was correct. Each time the user changes a quantity, the model registers a signal that some variable was not fully captured. Over repeated ordering cycles, the system begins to understand the habits of the individual shop owner.

Consider a shop that sells many bottles of Coca-Cola during a sports broadcast in the neighborhood. The AI does not automatically know about the event, but it notices that unusually high sales took place on the previous weekend. When a similar event appears on the calendar next time, the system can recommend a slightly larger order. Likewise, the system can detect a slow-moving flavor and recommend a smaller amount or a promotional pack to help clear the inventory.

Retailers also become stakeholders in the process. If they consistently adjust the algorithm's advice in a particular direction, their behavior teaches the system to adapt. This is a different approach from static software, which would apply the same rule to every store. With AI, every outlet can gradually get a personalized ordering model that reflects its unique mix of customers, physical space and competitive environment.

The human role in an AI-assisted supply chain

Introducing AI into retail ordering does not mean people disappear from the process. Field sales representatives, delivery drivers and merchandisers remain essential. What changes is the nature of their work. In the past, a sales representative might need to count every bottle in the storeroom, record numbers by hand, and then attempt to convince the retailer to place a larger order. That routine task took up time and left less space for building relationships or improving shelf displays.

With AI providing the underlying order recommendation, the sales representative can focus on high-value consultations. They can explain why a certain product should be placed at eye level, suggest a new flavor that consumers may want to try, or arrange in-store marketing materials that draw attention to a particular range. The retailer is treated less like a passive receiver of an order and more like a business partner who can make informed decisions.

Delivery drivers are also affected in positive ways. Because the AI system creates more accurate orders, drivers find fewer returned products and fewer rejected parcels. That reduces the physical burden of unloading items that the store does not really need. The driver also benefits from better route planning. When orders are known in advance, each stop is clearer, the vehicle can be loaded in reverse order of the deliveries, and time at each location can be kept to a minimum.

At the warehouse, managers can plan labor resources more precisely. Instead of predicting the volume of orders only from past averages, they can use AI forecasts for each day of the week. The staff can be deployed to match expected activity, reducing idle time and overtime costs. This kind of coordination is possible only when the ordering process at the end of the supply chain is digitized and connected to the upstream planning system.

Wider implications for beverage distribution

Coca-Cola's move is part of a larger trend in the food and beverage industry. Many manufacturers are looking for ways to make the traditional retail channel more visible and data-rich. The modern route-to-market has become competitive because consumer goods companies are under pressure to reduce waste, optimize logistics and respond quickly to changing tastes. AI is one of the most promising tools for meeting those expectations.

For Malaysia, the potential payoff is high. A more intelligent supply chain means less food waste and fewer commercial losses. It also means that smaller businesses are not excluded from the digital transformation sweeping through the economy. They can use tools that were once reserved for giant companies, and they can enjoy benefits such as better availability of products, fresher stocks, and lower risk of overpaying for goods that may expire before they are sold.

The system could eventually evolve to include more services. Once Coca-Cola and the retailer have an accurate digital record of orders, it becomes possible to support access to microcredit or digital payment solutions. Retailers may be able to use their order history as evidence of business performance when applying for financing. The AI could also connect to equipment monitoring in coolers and vending machines, creating an even more seamless replenishment experience.

Another important implication is environmental efficiency. When orders are accurate, delivery routes are better planned and fewer trucks need to make unnecessary trips. Less packaging is wasted on products that go past their shelf life. These contributions may appear small at the level of a single shop, but across hundreds or thousands of outlets they create a meaningful reduction in the carbon footprint of the distribution network.

From a market perspective, Coca-Cola is strengthening the long-term competitiveness of physical retail in Malaysia. E-commerce is growing, but the convenience of buying a cold drink at a nearby shop remains deeply embedded in Malaysian daily life. By making it easier for that small shop to stay fully stocked and financially healthy, Coca-Cola is also reinforcing the relevance of local retailers in an increasingly digital economy.

In the end, the effective use of AI is not about replacing judgment. It is about supporting people with better information so that they can make better decisions. For the Malaysian retailer ordering a case of Coke, the AI serves as a silent partner that remembers every busy weekend, every quiet weekday and every holiday spike. For Coca-Cola, that same data flows back into a system that manufactures, ships and sells the beverages more intelligently. The first obvious outcome is a smoother ordering process. The longer-term opportunity is a modern, responsive and resilient supply chain that benefits the company, the retailer and the consumer alike.

Source: AI News News

Speaking to reporters on the eve of the third Test, Babar made it clear that the sweeping changes were not his idea and that he learned about them only after they had been made. His remarks are the first public indication that the PCB's aggressive overhaul of the touring party was carried out without the captain's involvement.

Babar's Revelation

When asked about the decision to release seven players following Pakistan's heavy defeats in Leeds and at Lord's, Babar did not hide his surprise. "It was news to me," he said. "I found out after. The PCB decided it, so they'll be able to explain better."

The Pakistan captain was then pressed on whether the decision had been discussed with him before the squad was trimmed. Again, he deflected responsibility firmly toward the PCB hierarchy. "The PCB's decision must have been taken after some thought process," he added. "It's better if you ask them why."

Those comments suggest a clear disconnect between Babar and the board led by Mohsin Naqvi, who has taken an increasingly hands-on approach to team management since assuming the PCB chairmanship. Babar's tone was measured but his words spoke volumes about the lack of communication between the dressing room and the boardroom during one of Pakistan's most turbulent tours in recent memory.

Seven Players Released

The PCB's mid-series overhaul was announced in the aftermath of Pakistan's 194-run defeat at Lord's. England's victory gave them an unassailable 2-0 series lead, and the board responded by releasing Mohammad Rizwan, Imam-ul-Haq, Salman Ali Agha, Ali Usman, Aamir Jamal, Muhammad Awais Zafar and Khurram Shahzad from the touring squad.

The decision came after a series of calamitous batting displays by Pakistan. In the opening Test in Leeds, England won by an innings and 103 runs after Pakistan collapsed in both innings. The pattern continued at Lord's, where Pakistan failed to build competitive totals and suffered another emphatic defeat. Across the first two Tests, Pakistan did not reach 200 in any of their four innings, raising serious concerns about the team's preparation and technique against England's attack.

The overhaul was not limited to playing personnel. The PCB also removed head coach Sarfaraz Ahmed and bowling coach Umar Gul from their positions, bringing in white-ball head coach Mike Hesson and Ashley Noffke to oversee the team for the Edgbaston Test. The mid-series change in coaching staff added another layer of instability to an already chaotic tour.

Updated Squad and Young Talent

The revised squad named by the PCB includes Babar Azam, Abdullah Fazal, Abdullah Shafique, Arafat Minhas, Azan Awais, Mohammad Abbas, Mohammad Ali, Muhammad Imran, Mohammad Imran Jnr, Muhammad Ghazi Ghori, Razaullah, Saad Baig, Saim Ayub, Sajid Khan, Saud Shakeel, Shan Masood and Ubaid Shah.

The new look squad features several relatively untested players, and Babar hinted that the influx of youth might bring fresh energy to a dressing room that has been under intense pressure since the start of the tour. "Any young player who comes in is excited, and that excitement is visible," Babar said. "I believe that will be visible in the match, too."

Babar stressed that the team had little to lose in the final Test, given that England had already wrapped up the series. "The series is lost but we have nothing to lose," he explained. "We should play freely and believe in ourselves as a team. We've practiced a couple of days and explained things to the new boys."

The Pakistan captain is hoping that the new faces in the squad will approach the game without any fear or baggage, allowing them to express themselves on the field. His comments reflect a desire to end what has been a difficult tour on a positive note, despite the backdrop of institutional turmoil.

Babar's Form and Defence

Babar has also faced questions about his own performances during the tour. His numbers in England have not matched the high standards he has set elsewhere in his career, leading to criticism from former players and fans alike. However, he was quick to point to his recent form in the Caribbean, where he was named Player of the Series.

In Pakistan's Test series against the West Indies, Babar scored 193 runs at an astonishing average of 96.50. That reminder of his class was clearly directed at those who have questioned his place in the team and his effectiveness as captain. While he has not had a memorable time in England with the bat, Babar remains the focal point of Pakistan's batting lineup and their hopes of avoiding a 3-0 series whitewash.

The third Test carries significance well beyond the scoreline. Pakistan have already surrendered the series, but a defeat at Edgbaston would mark a particularly low point in the team's history. For Babar, leading a heavily altered squad in such circumstances will test his leadership skills to the limit. His public uncertainty about the PCB's decision has also raised questions about his future as captain, although he has not directly addressed that speculation.

The Bigger Picture

This series has been overshadowed by off-field drama, with persistent rumours of friction between Babar and Mohsin Naqvi. The PCB chairman has overseen sweeping changes since taking office, including regular alterations to the coaching staff and selection panel. Many cricket observers believe that the decision to overhaul the squad midway through a crucial series was made without sufficient involvement from the captain, and Babar's latest comments will only strengthen those claims.

The release of senior players such as Mohammad Rizwan and Imam-ul-Haq, both of whom have substantial Test experience, has been particularly controversial. Rizwan, in particular, is regarded as one of Pakistan's most dependable wicketkeeper-batters, and his absence from the final Test has puzzled many experts. The board has not issued a detailed explanation for why these specific players were released, which has fuelled further speculation about dressing room divisions and political interference in team selection.

Mike Hesson, who has been brought in to guide the side for the final Test, faces a daunting task in trying to steady a team that is clearly low on confidence. The partnership between Hesson and Ashley Noffke is expected to bring a new tactical approach, but with only a few days to prepare, their impact may be limited. Babar acknowledged that practice sessions have been focused on helping the newcomers understand their roles and adapt to the Test environment.

Looking Ahead to Edgbaston

The third Test at Edgbaston is scheduled to begin on September 9 and run until September 13. A green-tinged pitch is expected to offer assistance to the bowlers, which could create further problems for a batting lineup that has already struggled against England's pace attack. However, Babar has urged his young players to embrace the challenge rather than shy away from it.

"We should play freely and believe in ourselves as a team," he reiterated. That message of positivity is likely to be central to Pakistan's approach in Birmingham. With nothing to lose and everything to gain, the visitors may finally play without the conservatism that has hampered them in the earlier matches.

The presence of young players such as Abdullah Fazal, Arafat Minhas and Muhammad Ghazi Ghori adds an element of unpredictability to Pakistan's plans. If these debutants can handle the pressure of a Test match against England in front of a packed crowd, they could provide the spark that the team has been missing throughout the series.

Babar's own form remains a major talking point. He has always carried the weight of expectation on his shoulders, but this tour has been particularly difficult. Still, his recent success in the West Indies suggests that his talent has not diminished, and a strong performance at Edgbaston would go some way toward silencing his critics.

For the PCB, the fallout from this series is likely to continue long after the final ball is bowled. Questions about the relationship between the board and the captain, the process behind the squad overhaul, and the long-term direction of Pakistan cricket will not disappear quickly. Babar's pointed comments have ensured that the spotlight remains firmly on Mohsin Naqvi and the PCB's decision-making, even as the team prepares for one more battle on the field.

As Pakistan take the field in Birmingham, the hope among their supporters is that the turmoil off the pitch will not overshadow the cricket itself. The players have a chance to restore some pride and earn a measure of respect in a series that has been defined by heavy defeats and administrative chaos. The final Test begins on September 9, and Pakistan will be eager to show that despite everything, they can still compete with one of the world's best teams.

Source: MSN News

The defeat was a fitting—though painful—encapsulation of a season that began with immense promise and ended with the 25-year-old holding just one WTA trophy. Świątek, the former World No. 1 and the 2022 US Open champion, will now look back on 2026 as the first year since 2021 in which she failed to capture a major singles title. It was a stat line that seemed unthinkable when she arrived in New York carrying the form that had produced a title in Toronto and a semifinal run in Cincinnati.

A Stunning Collapse

Świątek’s path through the early rounds in New York had been, by her standards, quite tidy. She dismissed Wang Xiyu 6-2, 6-3 in the first round, eased past Nadia Podoroska 6-3, 6-2 in the second, and then endured a high-pressure battle against Marie Bouzková in the third round, eventually scraping through 7-6(3), 7-6(3). The Pole seemed to be building momentum with each match, and her powerful baseline game looked well suited to the punishing courts of the Billie Jean King National Tennis Center.

But Zheng Qinwen, a qualifier who had once been ranked as high as No. 7 in the world, was no ordinary opponent. The 23-year-old had suffered through a brutal period marked by a lingering elbow injury that ultimately required surgery, sending her tumbling out of the top 100. Coming into the US Open, she had to navigate qualifying just to reach the main draw. Once there, she displayed the kind of grit and shot-making that had once made her a Grand Slam finalist. Zheng opened with three-set victories over Kristina Liutova and Yulia Putintseva, then produced one of the signature wins of the tournament by rallying from 0-5 in the deciding set against Madison Keys, the 2025 Australian Open champion, to win 1-6, 7-6(3), 7-5.

Against Świątek, Zheng again found herself trailing by an alarming margin in the opening set. Świątek came out firing, breaking Zheng multiple times to race to a 5-0 lead with a double-break in hand. The crowd, which had already witnessed a similar comeback in Zheng’s previous match, began to sense that nothing was truly decided. Sure enough, Świątek’s level dipped. Her trademark forehand, usually a laser-guided weapon, started spraying errors. Zheng, by contrast, began reading her opponent’s serve better and stepping inside the court to take time away from the Pole.

“I felt like I started missing too many shots. She started playing in. I let her back into the match,” Świątek admitted in her post-match press conference. The statistics backed that assessment: Świątek’s first-serve percentage dipped below fifty percent in the opening set, and her unforced error count ballooned. Zheng broke back and, after leveling at 5-5, seized the momentum. She took the set 7-5, and the wind appeared to go out of Świątek’s sails. The second set followed a similar pattern, though without the early lead. Zheng broke serve early and never looked back, closing out the match with a confident 6-3 set to seal her place in the quarterfinals.

Press Conference Tension

The questions that followed Świątek into the interview room were perhaps more pointed than usual, and the former World No. 1 seemed in no mood to dwell on what many considered a catastrophic collapse. When a reporter asked how she processed such an unusually lopsided scoreline that had slipped away, Świątek gave a blunt and sarcastic reply.

“It’s just one match. I’ll try to play the same way as usual. What do you mean? Like, do you think it’s something crazy?” she said, her tone making clear that she found the question to be somewhat exaggerated. “You know, you can lose a set from any score. So, nothing special about that. Just that’s what happened, and that’s it.”

Her response drew a few raised eyebrows in the room, but it was consistent with the stoic, almost unbothered public persona that Świątek has often presented in difficult moments. Still, it was difficult to ignore the tension in her voice. When another reporter mentioned the often-chaotic nature of the US Open and asked whether that had affected her comfort level, Świątek was equally curt.

“I think it’s chaotic for everyone. So you just need to deal with it, and I’ve been fine. Thanks.”

She did, however, offer a brief explanation for the medical timeout she took during the third round match against Bouzková, which had raised concerns among fans about her physical condition. The issue, she said, was a tired quadriceps muscle that required a quick treatment from the physio.

“I’ve been fresher, but just little, you know, my quad is quite tired, but it had no influence on the game. Just wanted to give it a little boost, but yeah, it didn’t influence anything.”

Zheng Qinwen’s Cinderella Story

While Świątek’s exit dominated the headlines, the other side of the net belonged to a player whose journey to this moment was arguably even more compelling. Zheng Qinwen had burst onto the scene in 2024, when she reached the Australian Open final and captured the tennis world’s attention with her powerful serve and fearless playing style. She also won gold at the 2024 Paris Olympics, cementing her status as one of the rising stars of the WTA tour. However, a persistent elbow problem soon derailed her trajectory, forcing her to undergo surgery and spend months on the sidelines. Her ranking, once a career-high of No. 7, fell outside the top 100, and she was forced to play qualifying at Flushing Meadows, a stage she had once graced as a seeded player.

But Zheng has never lacked self-belief. After qualifying, she disposed of her first two main-draw opponents in three sets each, grinding out wins against Kristina Liutova and Yulia Putintseva. Then came the third-round encounter with Madison Keys, where Zheng once looked utterly outclassed. She lost the first set 1-6 and fell behind 0-5 in the third set. It was then that the Chinese star found a level that stunned the stadium: she won seven consecutive games to complete one of the most improbable comebacks of the tournament, closing out the match 1-6, 7-6(3), 7-5.

Against Świątek, Zheng again appeared on the brink in the opening set at 0-5, but she did not panic. Instead, she leaned on her experience and her conditioning, steering Świątek into longer rallies and forcing the Pole to go for more than she was comfortable with. The result was a masterclass in pressure management. Zheng’s victory not only earned her a place in the US Open quarterfinals but also served as a powerful statement that she had returned to the game’s elite level. For a player who had dropped so far in the rankings, it was a dramatic redemption arc.

A Season of Missed Opportunities

For Świątek, the US Open loss represented the final chapter of a 2026 major season that had started with so much hope. At the Australian Open, she reached the semifinals before falling in a tense three-setter. At Roland Garros, where she has historically been nearly unbeatable, she was stunned in the quarterfinals—a result that still ranked among the most shocking upsets of the year. Wimbledon, a tournament she has often struggled to master, saw her exit in the fourth round. Combined with the fourth-round defeat in New York, Świątek won just 12 matches in Grand Slam play in 2026, her lowest tally since the year she won her first major title in 2020.

The statistic is all the more striking given that Świątek has been a dominant force on the WTA circuit for the better part of half a decade. She has spent many weeks at the top of the rankings, claimed numerous WTA 1000 titles, and established herself as the most consistent clay-court player of her generation. Yet on hard courts, particularly at the US Open, she has often appeared a step slower or a shot less reliable than her rivals. She had reached the quarterfinals in New York in each of the previous two years, and many analysts believed that with her improved serve and more aggressive mindset, she was finally ready to challenge for a second title at the event.

The American swing did offer some encouragement. Świątek won the WTA 1000 event in Toronto, defeating several top-20 players along the way. She then advanced to the semifinals in Cincinnati, where she lost narrowly to the defending champion. These results were enough to place her among the pre-tournament favorites in New York, and her early-round form only reinforced that impression. Yet in her fourth-round match against Zheng, the old vulnerabilities resurfaced: a lapse in focus, a dramatic dip in first-serve percentage, and an inability to close out a set she had thoroughly dominated. The 5-0 collapse was not an anomaly; it was, as one commentator put it, the worst possible scenario for a player trying to prove she could be trusted at the majors.

What’s Next for Świątek

The immediate future for Świątek involves turning her attention to the Asian hard-court swing, a segment of the season where she has traditionally collected significant points. She is scheduled to play at a WTA 500 event in Guangzhou, followed by the WTA 1000 tournament in Beijing, where she has enjoyed strong support from the Chinese fans. These results will be crucial for her ambitions of qualifying for the WTA Finals, which will be held in Fort Worth, Texas later this year. With her rivals accumulating points at the majors, Świątek needs to reach the latter stages of several more tournaments to secure her spot in the year-end championship.

But the most pressing concern remains mental rather than technical. Świątek has often spoken about the importance of staying present on court, yet her body language during the match against suggested she struggled to recover from the shock of losing a 5-0 lead. The question that she dismissed with such irritation is the same one that tennis observers have repeatedly asked over the years: how can a player of her caliber, with so many titles and such deep experience, repeatedly find herself vulnerable to momentum swings of this magnitude?

Świątek’s response—that losing a set from any score is common—was technically correct. Players have blown big leads before, and tennis is famously a game of momentum shifts. But the manner of this defeat, coming at a Grand Slam against a qualifier who had to rally from 0-5 in her previous match, suggests that there may be deeper issues to address. The Pole does not lack for confidence, nor does she lack for skill. What she seems to be searching for is a way to bridge the gap between her often flawless groundstrokes and the erratic decision-making that surfaces when pressure reaches its peak.

With the Asian swing beginning in a matter of days, Świątek will have an opportunity to reset and rebuild her rhythm. The WTA Finals remain within reach, and a string of deep runs could still make this season look respectable from a points perspective. But no amount of wins in smaller tournaments will erase the sting of this US Open exit if she fails to address the pattern that produced it. For now, her fans can only hope that her blunt answer to a lingering question will not prove to be the most memorable moment of her 2026 season.

Source: AOL.com News

Diljit’s association with the phrase “I’m an Artist Bro” dates back to his response to a question about the agitation during a July Instagram Live session. When asked about the CJP protest, the singer had said, “Keep me away from all this… Bro, I’m an artist. I’m no politician.” Those words, recorded on a live stream, soon became a talking point. Some criticised him for not engaging with the students’ cause, while others defended him by saying entertainers should not be forced to take political positions. Now, with the release of “Devil”, those same words are being read in a different context.

What “Devil” shows in its visuals

The music video opens with an atmosphere of tension. It features large crowds of young people, many of them carrying signs and banners. The placards displayed in the video carry messages such as “Save Our Future” and “Our Future, Our Fight”. Police personnel are also shown, and there are sequences depicting clashes between protesters and authorities. The video does not mention the CJP by name, nor does it refer directly to Jantar Mantar. Yet the visual language has led viewers to connect it with the student protest that unfolded in Delhi over the past few months.

One of the most discussed scenes in the video involves a little girl walking toward a police officer. She holds a white rose in her hand. The officer, after a moment, approaches her and embraces the child. This brief sequence has been interpreted by many fans as a plea for peace and dialogue amid high tension. It also stands in sharp contrast to the more aggressive scenes that appear earlier in the video. The message, as many viewers understand it, seems to be about the need to protect the future and to listen to the voices of the young.

Social media users have flooded comment sections with their readings of the video. One viewer wrote, “Peak detailing video!!!!!! Student protest, Our Future our rights.” Another said, “iykyk this video is about which protest only Diljit can do this.” The phrase “iykyk”, which means “if you know, you know”, is frequently used by fans to suggest a hidden reference that does not require explicit confirmation.

Inside “I’m an Artist Bro” and its seven tracks

“Devil” is one of seven tracks on Diljit’s latest album, “I’m an Artist Bro”. The project also includes the songs “Fizawan”, “Love Like This”, “Gustakhi”, “Waliyan”, “Sweet Ting” and “Lambo”. The title track has become a catchphrase in its own right, especially after Diljit’s previous statements about staying away from politics. The album as a whole blends upbeat Punjabi pop with more serious themes, and “Devil” stands out because of its overt cinematic treatment of protest and public unrest.

From “I am no politician” to a protest-themed project

The title “I’m an Artist Bro” carries a special significance in Diljit’s recent career. It is the same statement he made in July, and it has now become the title of his latest album. The video for “Devil” ends with the disclaimer “Don’t try this at home”, followed by the words “I’m an Artist Bro.” This has been seen by fans as a deliberate callback to the controversy that followed his Instagram Live comments.

The singer, who has a massive following in Punjab and beyond, has often been careful about taking public positions on politically sensitive issues. After the farmers’ agitation of 2020-21, when he had voiced support for the protestors, he experienced both praise and backlash. His July remarks about the CJP-led protest added another layer of complexity. Many young fans wanted him to speak up, but he initially chose to remain indifferent, saying that he was not aware of the details of the protest and preferred to focus on his work.

Later, however, Diljit did make a statement about the students at Jantar Mantar. He said that the students should not have been treated in that way and asked authorities to listen to their demands. He also wrote, “The voice of the people is the voice of God.” These remarks, while not as expansive as some had hoped, were noted by his supporters. The new video, according to many viewers, appears to align with that later sentiment rather than with his earlier caution.

Why the comparisons have gained traction

Several elements in the “Devil” video have reinforced the comparisons with the CJP protest. The setting is clearly a public demonstration, with hordes of youths occupying the frame. The signs carried by the protesters highlight fears about education, employment and the future. Such themes have been central to the discourse around the Jantar Mantar agitation, where students have been demanding answers on a range of issues related to youth welfare and opportunities.

The video also includes members of the Sikh community among the demonstrators, which further strengthens the link for viewers. Diljit, being a prominent Sikh artist, often includes culturally specific imagery in his work. In this instance, the presence of Sikh men in turbans and young women in vibrant attire adds a localised texture to the protest scene. The combination of colours, placards and police lines makes the video feel like a dramatised recreation of a campus protest.

There is, however, no official confirmation from Diljit or his team that “Devil” is about the CJP protest. The video makes no direct reference to the organisation, its leadership, or the specific demands put forward by the students. The interpretation is being driven largely by the audience. The return of the phrase “I’m an Artist Bro” at the end of the video has added to the speculation, because that phrase itself became a part of the political conversation around Diljit’s silence.

Reactions from fans and critics

The reactions to the video have been mixed, but the dominant sentiment among Diljit’s fans is one of pride. Many feel that the singer has subtly responded to the criticism he faced after his July comments. One fan commented, “He stand proudly with the students but the main motive of this video was to promote peace rather than wars!! the og ARTIST.” Another follower wrote, “This video is a slap on those people’s face who mocked Diljit for not standing with students in protest. He is an artist but was already in a controversy once due to farmers protest few years back and if he said anything on the protest haters would have criticised him there too. Government and all those people who criticised him should give him an apology.”

A separate comment read, “A banger song, video concept, a deep message.” Another user described Diljit as the “Biggest punjabi artist on the planet FOR A REASON.” However, not all responses were filled with praise. A smaller section of social media users questioned why the singer had chosen to express himself through a music video rather than through a direct statement. One post said that if Diljit really wanted to support the students, he should say so clearly instead of packaging the message in art.

There were also those who interpreted the video as an indirect criticism of the security forces. One commenter wrote, “Bro’s indirectly taunting Indian forces came at the students protest to use lethal force on students. He is a legendary artist and screw the ones who were making fun of him.” This kind of reaction underscores the deeply polarised nature of the online discourse around both Diljit and the CJP agitation.

Art, ambiguity and public expectation

The debate around “Devil” raises a larger question about the role of artists in times of protest. Should a celebrity be expected to make political statements? Does a music video with political imagery count as a form of engagement? For many of Diljit’s supporters, art itself is a vehicle for commentary, and “Devil” is no exception. They argue that the singer has chosen a language that allows him to express solidarity without limiting himself to a single political slogan.

Diljit’s position in the entertainment industry is important in understanding the weight of these expectations. Over the past decade, he has gained international recognition for his work in Punjabi and Hindi cinema, and he is known for his electrifying live performances. His fanbase spans extensive geographies, including India, Canada, the United Kingdom and Australia. Every move he makes is scrutinised, and his statements on sensitive issues are often amplified beyond his intent. That was true during the farmers protest, and it is true again now.

The singer has frequently avoided direct political labels. He has, at times, supported causes such as equality and justice, but he has also insisted that his primary duty is to entertain. The “Devil” video, however, has blurred that line for many viewers. By choosing to fill a music video with protest imagery, Diljit has opened the door to multiple interpretations. Some see him as having taken a stand, while others see the video as a broader statement about the abuse of state power and the vulnerability of young people.

No official word yet

As of now, Diljit has not clarified the meaning behind “Devil”. Neither his social media accounts nor his music label have released a statement explaining whether the video was intended to reference the CJP protest. The absence of an official explanation leaves the ambiguity intact. Fans are free to project their own interpretations onto the visuals, and both those who praise the video and those who criticise it can claim some support from the imagery.

What remains clear is that “Devil” has succeeded in generating conversation. It has also brought the phrase “I’m an Artist Bro” back into the spotlight. What was once a simple response to a question has now become a slogan attached to an album and understood as a coding device for political commentary. Whether the song is a direct response to the student protest, a clever marketing strategy, or simply an artistic expression of conflict in society is still open to debate.

Until Diljit himself clarifies, fans and critics will continue to read between the lines. In a digital ecosystem where every visual detail is examined and every lyric is interpreted, “Devil” has already achieved something many songs do not: it has made people talk. For an artist who insisted he was not a politician, that is no small achievement.

Source: MSN News

Source: Coindesk News

Eight years of clipboard hijacking

The technique is known as clipboard hijacking or address replacement. It exploits a routine activity that almost every cryptocurrency user performs: copying a long alphanumeric address from an invoice, web page, chat message, or email and pasting it into a payment screen. Sality monitored the clipboard in real time. When it detected a pattern that looked like a Bitcoin or Ethereum address, it substituted a different address that had been prepared by the operator. The substitute was often structured to preserve the opening characters and sometimes the ending characters, so a victim who compared only part of the displayed address would see no obvious difference.

Most desktop wallets show the pasted address in a small box. A user may check the first few letters and the final few numbers, but the middle section can be ignored. The malicious address was selected from a pool of valid wallet addresses that the attacker controlled. When the victim confirmed the transaction, the digital asset left their wallet and traveled directly to the criminal's address. Because blockchain transactions cannot be reversed, the theft would only be discovered after the funds were already moved through exchanges or mixing services.

Despite the elegance of the scheme, the financial toll was relatively modest. Investigators linked Sality to at least 12.1 million rubles in crypto losses, which is about $150,000 over the full period. The figure likely undercounts the true total, because victims may not have reported all incidents or because some transactions were denominated in untracked cryptocurrencies. The small average payment may have helped the operation stay under the radar: it functioned as a long-tail scam rather than a single enormous hack.

Old malware with a new criminal module

Although Sality appeared in recent reports as a crypto botnet, its origins go back much further. Security researchers have tracked various versions of Sality since the early 2000s. The malware was classified as a file infector, meaning it could attach parts of its code to executable files.

Source: Coindesk News

Pons fee milestone signals launchpad shift

Data from DefiLlama shows Pons earned about $5.95 million in fees over a 24-hour period. That total placed it fourth among all services tracked by the dashboard, ahead of several protocols that had historically generated larger daily fee streams. The ranking put the memecoin maker above Pump, which pioneered simplified token launches on Solana, and above Hyperliquid, a derivatives-first ecosystem known for high trading volume. It even surpassed Robinhood Chain's own fee intake over the same stretch, despite the fact that every transaction on Pons ultimately settles on the network.

The record numbers reflect a broader shift in how retail users are entering crypto in the current cycle. Rather than buying major tokens on centralized exchanges, many newer participants are moving directly to launchpads to create or trade short-lived meme assets. These platforms can generate large aggregate revenue through high-frequency trading and creation fees, even when individual transactions are relatively cheap. Pons has become an illustration of that model in action, functioning more like a high-throughput marketplace than a conventional exchange.

Fee mechanisms on launchpads typically work at several levels. Token creation itself carries a set cost, and every subsequent swap or trade can include a protocol fee. On Pons, these fees are paid by users who are trying to get into new meme tokens early. The costs are low enough to encourage thousands of attempts, but the sheer number of attempts adds up quickly. When 25,000 tokens are launched in a single day, even a modest per-token fee becomes a substantial source of revenue.

Token launches and trading volume spike

At the center of Pons's fee surge is a burst of token creation. Nearly 25,000 new tokens were launched on the protocol on Sept. 2 alone, and daily trading volume on those and existing tokens reached $544 million. That pace of minting is comparable to some of the busiest periods seen on other memecoin platforms during their own growth spurts. The high number of new projects signals strong demand for instant token issuance, a format that appeals to traders hoping to discover the next viral asset before the crowd.

Most of the tokens created on Pons are not intended to become long-term projects. They are often launched with whimsical names or images, quickly bought and sold by traders hunting for early entries. Only a small fraction will achieve meaningful liquidity or survive more than a few days. But the aggregate volume is enormous, and every failed or forgotten token still contributes fees while being traded. This structural feature makes launchpad revenues less reliant on any one token's success and more dependent on the overall throughput of the platform.

Because Pons operates on Robinhood Chain, transaction costs are relatively low compared to older networks like Ethereum. The architecture also allows token deployment without needing to create a liquidity pool or configure complex parameters. Users can name a token, add an image and launch it almost instantly. That simplicity lowers barriers to entry, drawing in a wider pool of participants who might not have created tokens on more technical platforms. As more users arrive, the platform feeds on its own network effect: more tokens lead to more trading opportunities, which attracts more traders, which produces more fees.

PONS token benefits from fee-driven buybacks

The native token of Pons, called PONS, has also been a major beneficiary of the surge. The token rose roughly 300% over a seven-day period, according to market data accompanying the fee report. That rally was driven in part by the protocol's use of its fee revenue to conduct buybacks and burns. When Pons generates fees, a portion of the earnings can be used to buy PONS on the open market and permanently remove it from circulation, creating deflationary pressure.

These buybacks are funded by protocol earnings, not by future promises, which distinguishes the token from many memecoin tokens that rely entirely on community enthusiasm. In a rising fee environment, buyback-and-burn programs can create a positive feedback loop: higher fees lead to more token burns, which reduces supply, which can push price upward if demand is steady. The recent PONS price increase is cited by market observers as evidence that this mechanism is starting to be reflected in valuations.

Of course, high volatility remains a feature of the PONS market. Tokens like PONS can quickly surrender gains if daily fees decline or if traders rotate into competing platforms. Still, the week's performance has turned PONS into one of the larger gainers among tokens associated with launchpad protocols, and it has raised the profile of Pons as a possible hub for future memecoin speculation.

Robinhood Chain collects record fees from Pons activity

Robinhood Chain also benefited directly from the Pons explosion. The network collected close to $4 million in a single day, largely due to settlement demand generated by Pons-related traffic. The exact figure amounts to roughly one-fifth of all fees that Robinhood Chain has accrued since its launch in July. Put another way, the network appears to be experiencing a fee concentration in which one application accounts for an outsized share of usage only weeks after going live.

This concentration is both an opportunity and a risk. If Pons continues to attract users, Robinhood Chain could build an early reputation as the home of memecoin activity, similar to how Solana positioned itself in earlier meme cycles. Rising base fees and user onboarding in turn could create an ecosystem feedback loop by attracting additional developers. However, the reliance on one fee-heavy application also makes Robinhood Chain's macro fee statistics sensitive to the whims of memecoin speculators. Meme cycles can fade quickly, and networks that depend too heavily on gaming-like activity can see their metrics fall just as fast.

There is also a broader lesson in the relationship between an application and its base network. In the current bull cycle, much of the value created by decentralized applications is still captured by the blockchain itself through gas fees. Pons flips that dynamic, at least for one day, by earning more at the application layer than the network earns from all activity. This supports the idea that protocol hierarchies are not fixed and that applications can build independent fee engines on top of shared infrastructure.

A new phase of the memecoin fee race

Pons entering the top five of DefiLlama's fee rankings alongside major lending protocols, decentralized exchanges and blockchain networks demonstrates that token launchpads can produce highly meaningful cash flows. For years, launchpad revenue was often dismissed as a minor side effect of novelty trading. The current data challenges that notion by showing a memecoin application competing with the biggest fee generators in crypto.

Among the comparisons noted in the data, Pons surpassed Pump in terms of fees generated. Pump, which helped popularize the modern fair-launch token format, still enjoys an active user base, but its fee volumes at this specific time were less than Pons. Hyperliquid, a platform that has drawn thousands of traders to its perpetuals and spot experience, was also eclipsed during the tracked day. Such comparisons are a snapshot rather than a permanent hierarchy. Fees move rapidly from one platform to another, and rankings change as new features and narratives emerge.

Another important detail is that more than 25,000 token launches in one day is not just a record for Pons but also a stress test. A memecoin platform that allows unlimited token creation can be vulnerable to spam, malicious token contracts and copycat attacks. The fact that Pons handled 25,000 launches and $544 million in volume suggests the underlying infrastructure is capable of supporting extreme throughput. It also means the platform must invest heavily in security and user protection to maintain trust in an environment where new assets can appear every second.

Market context and broader implications

The current cycle of memecoin development has evolved from the earliest jokes to a more complex ecosystem involving multiple chains, launchpad aggregators, social tokens and protocol-owned liquidity pools. Pons is one illustration of this complexity. In a rapidly moving environment, branding and user experience matter more than underlying chain. A simple way for a non-custodial wallet to create a token in seconds is enough to attract high volumes.

Chain comparison tools that list fee metrics often rank network-level fees and protocol-level fees in different categories. Pons had more daily fees than reference protocols because it charges both for minting and for swapping, making it more similar to an application with its own order flow than to a pure network. Across the industry, this is part of a larger application-layer value capture thesis, where applications can capture fees that exceed the network fee base.

Fee aggregation platforms have also introduced a note of caution about daily rankings. A single viral token, a liquidity event or a promotional campaign can skew one day's numbers. Similarly, apps can temporarily increase fees during periods of high volatility, only to see those fees retreat when calm returns. But in Pons's case, the volume and launch counts are broad enough to suggest sustained behavior rather than an isolated spike. A protocol that creates tens of thousands of tokens in a day is unlikely to revert to zero activity overnight.

For users, the Pons phenomenon highlights the ongoing appeal of instant, low-cost token issuance and the willingness of some traders to pay substantial fees for the chance to catch a token before it goes viral. Whether this behavior represents a broader cultural shift or another cycle of speculative excess remains open to debate, but the fee numbers offer a clear baseline for the scale of memecoin appetite in the current market environment.

Source: Coindesk News

OpenAI’s announcement describes a model that goes beyond helping human hackers write better phishing emails or analyze source code. Astra, according to the company, can operate independently across the entire attack chain: finding a vulnerability, crafting an exploit, deploying it and achieving a real security impact. That level of autonomy raises hard questions about the safety of releasing such powerful digital tools into a world already struggling with ransomware gangs, state-sponsored attackers and widely deployed vulnerable software.

Key facts about Astra

• OpenAI’s upcoming Astra model can autonomously discover previously unknown software flaws and turn them into working attacks, earning the company’s first “Critical” cyber capability rating.
• In OpenAI’s own testing, Astra exploited known vulnerabilities, found two new flaws, escaped a hardened browser sandbox and combined operating-system weaknesses to gain root access.
• OpenAI says it has delayed parts of Astra’s development to add safeguards, and it plans to restrict the model’s most advanced cybersecurity features to selected external testers.
• Security researchers are concerned that tools like Astra could rapidly exploit flaws in cryptocurrency software, adding risk to exchanges, wallets and smart contracts.

What makes Astra different

The “Critical” label is not a marketing phrase. OpenAI has a well-known internal preparedness framework designed to assess models that might pose dangerous capabilities. Cybersecurity is one of the key risk categories in that framework, along with other areas such as chemical or biological knowledge. On the cybersecurity scale, models are evaluated for whether they can complete offensive tasks with increasing autonomy and sophistication. Reaching the Critical tier means OpenAI believes Astra can handle high-complexity vulnerabilities in difficult environments with minimal supervision, which puts it in a different class from earlier language models and agent systems.

The practical implication is enormous. Many existing AI security assistants are trained to explain code, suggest possible mitigations or match known vulnerability patterns. Astra, by contrast, is described as a model that can reason through an unfamiliar system, identify its logical weaknesses and build a reliable attack. That is the kind of skill that expert penetration testers spend years learning. If such a capability is successfully contained, it could be used to find and fix critical flaws. If leaked, stolen or deliberately misused, it could give attackers a force multiplier unlike almost anything in the modern history of software security.

Security researchers who have followed the development of frontier AI say this was always considered a dangerous endpoint. In previous years, large language models struggled to execute long tasks or use computer tools consistently. Today’s agents can navigate websites, write files, run terminal commands and correct their own mistakes. Combining those abilities with offensive security knowledge points toward a future in which AI models act as their own attackers, operating far faster than human analysts and learning from every successful or failed attempt.

What Astra’s testing actually showed

OpenAI described controlled testing in which Astra was given a computing environment and asked to find weaknesses. In one set of tests, the model successfully exploited known vulnerabilities without requiring a human to guide each step. It also discovered two flaws that had not been previously known, a particularly striking outcome because unknown vulnerabilities, often called zero-days, are usually among the most expensive and carefully guarded secrets in cybersecurity. Finding a zero-day routinely takes elite human experts considerable time; an AI that can do so quickly suggests that the barrier to discovering new attack paths may soon be much lower.

Even more concerning to the testers was that Astra managed to escape from a hardened browser sandbox. Browser sandboxes are one of the most important defensive technologies on the modern internet, because they isolate untrusted code such as JavaScript on a webpage from the rest of a user’s computer. Escaping a sandbox typically means an attacker can read personal files, access system memory and interact with other processes running on the same device. Astra reportedly did this, and then in a separate chain of actions, it combined multiple weaknesses in the operating system to escalate its privileges all the way to root access. Root access gives a program complete and unfettered control over a machine, allowing an attacker to install persistent malware, disable security software and move laterally across a network.

The fact that Astra could combine separate vulnerabilities into a chained exploit is especially meaningful. Real-world cyberattacks rarely rely on only one weakness. Attack chains often require a first flaw to gain a small foothold, a second flaw to leap to a more powerful process and a third to reach critical data or privileges. Automating that multi-step reasoning has been a target for security researchers for years, but most automated tools remain heavily guided by humans and perform poorly outside known playbooks. If Astra can independently create such chains, it moves closer to the flexible, adaptive attack capability once imagined only for elite hacker teams.

Why OpenAI is slowing down

OpenAI said it has delayed parts of Astra’s development rollout while the company adds safeguards. In practice, this means the model will not be offered with its full range of cyber capabilities to every customer, at least at first. Instead, OpenAI intends to place the most advanced features behind stricter gates, granting access to selected vetted testers who are equipped to handle the dangers of live vulnerability discovery and exploitation. The company is also reviewing the model’s training procedures to reduce the chance that it can be prompted to carry out unauthorized attacks once deployed.

This decision reflects a broader tension inside the AI industry. Developers face enormous commercial pressure to ship products that can handle difficult, multi-step tasks. A model that can navigate a computer like a human is valuable precisely because it is autonomous. Yet cyber capability is a dual-use trait: the same talent that lets a model secure a system also lets it attack another system. By delaying and restriction the most dangerous capabilities, OpenAI is acknowledging that competitive speed cannot be the only factor determining release timing.

Some policy analysts view the announcement as an important precedent. OpenAI’s internal scale now says that a Critical cyber model is not a hypothetical future risk but a current operational reality. That could push other AI developers to adopt similar evaluations, and it could push governments to consider regulation around autonomous cyber offense. A model with this strength could potentially sit outside traditional vulnerability disclosure systems because it finds flaws not in a specific product but in general classes of software. No existing regulatory process can reliably manage that kind of distributed discovery and exploitation capability.

Critical implications for cryptocurrency

No part of the technology industry is watching OpenAI’s warning more closely than the cryptocurrency ecosystem. Digital asset businesses rely on code that moves value instantly and without intermediaries, and many platforms are under constant threat from attackers driven by the possibility of anonymous financial gain. If a model like Astra can find vulnerabilities and write working exploits in minutes, the risk to smart contracts, DeFi platforms and cross-chain bridges could rise dramatically.

The concern is not based on abstract fears. Decentralized finance has already lost billions of dollars in hacks that exploited small logic errors, flawed authorization checks or rushed upgrades. Many of those attacks were conducted by humans who spent days studying protocol code and transaction flows before striking. An autonomous AI model with Astra’s demonstrated skills could shorten that process to a matter of hours or even minutes. Every public contract becomes a more attractive target because the AI can rapidly scan for vulnerable patterns and immediately craft a transaction that drains funds.

Cryptocurrency software is an especially appealing target for advanced AI because so much of it is open source. Public code is easy for a model to analyze, and blockchains themselves provide a transparent ledger that reveals exactly where money is stored and how it moves. An attacker can search for misconfigured protocols, reward functions, exchange integrations or upgradeable contract patterns across hundreds of projects in a single attempt. Add instant liquidity and pseudonymous transfers, and the result is a security environment in which autonomous attacks could be both extremely fast and very difficult to reverse once launched.

The same tools that cause concern for crypto defenders could eventually be used for defense. AI models trained on vulnerability discovery might help auditors review smart contracts before deployment, locate reentrancy bugs, and verify that patches are effective. However, the current announcement makes clear that OpenAI is still wrestling with how to prevent the offensive side from outpacing the defensive side. Limiting Astra’s capabilities to trusted testers may slow the spread, but it does not eliminate the possibility that similar methods emerge from other labs or from malicious actors who obtain access.

Autonomous cyber capabilities are accelerating

OpenAI’s evaluation of Astra is part of a broader transformation in cyber tooling. Several years ago, AI models could only generate snippets of code or help attackers fine-tune phishing messages. Then came models that learned to operate browsers, launch command-line tools and automate penetration testing workflows. Today, frontier agents are expected to complete long chains of reasoning, test new ideas and revise their actions in real time. Cybersecurity has become a critical benchmark for this type of reasoning because successful attacks require planning, adaptation and deep knowledge of cause and effect.

Governments have also noticed. Several countries have warned that AI could reduce the skill barrier for offensive cyber operations, enabling smaller groups to carry out attacks once reserved for nation-states. At the same time, state-backed security agencies see autonomous AI as a way to discover vulnerabilities in critical infrastructure before enemies do. That dual-use reality makes the public release of any model with a Critical cyber rating an event with geopolitical consequences. OpenAI’s report on Astra may contribute to a future in which AI models are controlled under strict licenses, monitored through real-time usage limits and required to pass safety tests before entering the market.

The deeper question posed by Astra is how much trust the security community should place in the idea of an AI that can attack without human help. Even if OpenAI keeps the model in the hands of trusted testers, the underlying research advances the state of the art across the entire field. Other labs may develop similar models with fewer safeguards. The result could be a wave of automated vulnerability discovery that outpaces the human security workforce responsible for patching software, hardening networks and responding to incidents.

For now, OpenAI’s announcement stands as a clear signal that the era of purely assisted AI hacking is ending. Astra does not need a human to hold its hand while it locates a flaw, writes an exploit and breaks through layers of defense. The practical responses—delays, safeguards, restricted access—buy time but do not stop the technology from evolving. The real question is no longer whether AI can build attacks on its own, but whether defenders can adapt at an equally fast pace.

Source: Coindesk News

CHARLOTTE, N.C. — September 3, 2026 — The Broadcast Retirement Network (BRN) today is marking seven (7) years of daily, advertising-free, fact- and evidence-based programming dedicated to retirement, aging, finance, lifestyle, privacy, and wellness. 

Since its launch in 2019, BRN has aired more than 2,500 original programs, broadcasting seven days a week at 7:30 AM ET. Over that period, the network has produced more than 625 hours of programming featuring international experts sharing their expertise with the audience. 

BRN programming is made available at no cost, complete with full transcripts, and is broadly syndicated across major news sites, aggregation services, streaming platforms and podcast services—ensuring BRN’s content reaches its audience wherever media is consumed. Shorter clips are also distributed across all major social media channels every two hours, further extending the network’s reach.

BRN distinguishes itself through a straightforward editorial commitment: no sales pitches, no advertisements, and no politics—just the facts, every morning. The network also delivers a daily, hand-curated newsletter The Morning Pulse on aging, finance, lifestyle, privacy, retirement, and wellness, selected by an expert editor.

“Reaching our seventh anniversary is a testament to the trust our audience and partners have placed in us,” said Jeffrey Snyder, Chief Executive Officer and Lead Anchor of the Broadcast Retirement Network. “For seven years—drawing on my 32 years in the retirement industry, our mission has remained the same: to deliver clear, credible and useful information to the people who need it, free of any external noise.

A September-Only Anniversary Offer

To mark the occasion, BRN is offering new subscribers 20% off The Morning Pulse—its daily newsletter delivering expert-curated news on money, health, and retirement, written by a human, without the use of AI, and free of ads and sales pitches. Every subscription directly supports BRN’s daily programming. Read today’s edition here: https://us6.campaign-archive.com/?u=26e6dd4c63255e9ef5e07a4c4&id=152e013f21

Normally $4 per month, new subscribers can save 20% with code BRN20 through September 30, 2026, only. Individuals can subscribe at https://buy.stripe.com/4gw9CS5NI8cibVm7su.

Looking Ahead

BRN will unveil a new opportunity for prospective partners on September 9, 2026. Details will be shared across the network’s platforms and social media channels.

“This milestone belongs to our guests, audience and partners as much as it does to us,” Snyder added. “We are grateful for seven years of continued support—and we’re just getting started.”

About Broadcast Retirement Network

The Broadcast Retirement Network (BRN) is an independent daily program delivering fact-based news and expert insight on aging, finance, lifestyle, privacy, retirement, and wellness. Airing seven days a week at 7:30 AM ET, BRN provides advertising-free programming, complete with transcripts, syndicated at no cost across major news, streaming, and podcast platforms. BRN is led by Chief Executive Officer and Lead Anchor Jeffrey Snyder, who brings 32 years of retirement industry experience to the network’s daily coverage.

Media Contact

Jeffrey Snyder 

Chief Executive Officer / Lead Anchor 

Email: jeff@broadcastretirementnetwork.com 

YouTube: https://www.youtube.com/@BroadcastRetirementNetwork

###

• Email: jeff@broadcastretirementnetwork.com
• Website: https://www.youtube.com/@BroadcastRetirementNetwork
• Company: Broadcast Retirement Network
• Company Logo:
• Name: JEFFREY H SNYDER
• Phone No: 6464840188
• City: Charlotte
• State: NC
• Country: United States

<p>The post Broadcast Retirement Network Marks 7 Years of Daily, Fact- and Evidence-Based Programming first appeared on PR Business News Wire.</p>