Chinese artificial intelligence startup Moonshot AI has sent shockwaves through the global tech sector with the launch of its latest breakthrough model, Kimi K3.

Led by 34-year-old entrepreneur Yang Zhilin, the Beijing-based company has ignited widespread anxiety across Western markets, triggering a sharp sell-off in global technology and chip stocks.

Kimi K3 boasts 2.8 trillion parameters, making it the largest open AI model built to date. Independent benchmarks indicate that its performance places it neck-and-neck with elite Western frontier models from OpenAI and Anthropic.

Crucially, the model features a massive one-million-token context window, an always-on reasoning “thinking mode” and powerful multimodal capabilities. Furthermore, Moonshot’s commitment to releasing the model’s open architecture allows businesses and developers worldwide to download, fine-tune, and self-host the technology at a fraction of closed-source API costs.

Global financial markets were instantly spooked by the release because Kimi K3 directly undercuts the staggering infrastructure and financial advantages long held by Silicon Valley giants. Investors fear that high-performance, cost-effective, open-source alternatives from China will dismantle the pricing power of US tech leaders and stall the astronomical hardware investments currently fuelling semiconductor powerhouses such as Nvidia and Samsung.

The sudden surge in demand for Kimi K3 even overwhelmed Moonshot’s infrastructure, forcing the company to temporarily pause new subscriptions. Nevertheless, the breakthrough has fundamentally altered perceptions of the global AI landscape, challenging long-held assumptions of Western technological supremacy.

The post Chinese startup Moonshot AI rattles global markets with Kimi K3 appeared first on Tech Digest.

Recent firmware discoveries indicate that the next-generation compact dish is officially entering the manufacturing pipeline, potentially bringing major improvements for off-grid users, campers and digital nomads.

The most significant upgrade over its predecessor is the inclusion of a built-in four-cell lithium battery pack managed by an advanced battery management system.

While the original Starlink Mini required an external power bank, DC cable, or AC adapter to function in the field, the Mini 2 operates as a truly self-contained unit, charging via USB-C.

Measuring 225 × 273 mm, the Mini 2 is slightly more compact than the first-generation model, while the average EIRP (Effective Isotropic Radiated Power) comes in at 27.4 dBW, fractionally below the Mini 1’s 27.57 dBW.

Hardware data also shows a reduced transmit duty cycle, dropping from 75% to 11%, which prioritises energy conservation and will likely result in slower sustained uplink speeds. However, download capabilities and GPS support for in-motion use remain fully intact.

SpaceX has not yet officially announced a formal release date or pricing structure. However, because the hardware has advanced to production-level firmware designations (mini2_prod1), industry watchers expect an official unveiling soon.

Given that the original Starlink Mini kit starts at $199, market analysts anticipate the battery-equipped Mini 2 will command a higher price tag to match its enhanced, all-in-one portability.

The post Starlink prepares to launch Mini 2 with integrated battery appeared first on Tech Digest.

Every autumn brings a new flagship chip promising double-digit performance gains over its predecessor. Manufacturers show off benchmark charts, and reviewers dutifully report the numbers. Yet ask most UK smartphone owners whether their banking app or social feed feels noticeably snappier than it did two years ago, and the answer is usually a shrug.

That disconnect isn’t imaginary. It reflects a genuine gap between what processors are capable of in laboratory conditions and what actually determines how fast an app feels in daily use. The bottleneck has shifted away from silicon and towards factors most marketing materials conveniently ignore.

Understanding why requires looking past the spec sheet entirely.

Why benchmark scores rarely match real-world speed

Benchmark tests measure raw computational throughput under ideal conditions: full battery, cool device, no background interference. Real-world app launches happen in messier circumstances, with dozens of processes competing for the same resources.

A chip that scores brilliantly in a synthetic test can still feel sluggish when the phone is juggling notifications, syncing photos, and running a keyboard’s predictive AI simultaneously. The processor isn’t the limiting factor in that scenario; contention for memory and thermal headroom is. This is why two phones with near-identical benchmark scores can produce noticeably different everyday experiences.

How app bloat quietly cancels out hardware gains

Average app sizes have grown steadily as developers bundle more features into every update. Embedded machine-learning models for photo editing, analytics software development kits, and increasingly complex animation frameworks all add weight. None of this is hidden or malicious; it simply means apps now ask more of the hardware than they did a few years ago.

Larger apps mean longer disk reads and heavier memory demands at launch, which can quietly erode the benefit of a faster chip. This pattern spans app categories, including entertainment platforms with rich graphics and real-time features, such as video streaming websites, travel apps, and poker platforms. Those interested in poker, blackjack, and similar activities can check on Gambling Insider what deals and features will attract players in the rest of 2026.

Cheaper handsets with slower storage feel this effect most acutely. A flagship processor can sprint, but if it’s dragging an increasingly heavy app behind it, the improvement barely registers to the person holding the phone.

Where network conditions outweigh processor improvements

For most apps that pull data from the internet, network conditions matter more than local processing power. Ofcom figures cited in the government’s ongoing mobile market review show mobile traffic across UK networks has risen sharply in recent years, adding pressure to infrastructure even as coverage expands. More users pulling more data through the same cells inevitably affects consistency.

Coverage maps also tell only part of the story. Research highlighted in a recent mobile quality report shows how widely daily mobile usage patterns vary across the UK, reinforcing that a phone’s chip is rarely the constraint once data has to travel across a congested or distant cell tower. A flagship processor waiting on a slow network connection is still a flagship processor waiting.

This matters because so many everyday apps, from messaging services to streaming platforms, are essentially thin clients pulling data from servers. The phone’s job is often just to render what arrives, not to compute anything demanding locally.

What actually moves the needle for users

If chip speed isn’t the primary lever, what is? Available RAM matters enormously, since it determines whether an app resumes instantly from memory or needs a slower cold start. Storage speed matters too, particularly on mid-range devices where cheaper NAND chips create bottlenecks that no processor upgrade can fix.

Network quality remains arguably the biggest variable of all. A recent data consumption report noted record levels of UK mobile data use, underlining just how much strain modern networks carry during peak hours. Until connectivity keeps pace with rising demand, faster chips alone won’t deliver the snappier experience consumers are promised each time a new handset launches.

The post Faster smartphones aren’t actually making apps load quicker appeared first on Tech Digest.

New Starlink firmware indicates that SpaceX has started producing a “Mini 2” dish, which features a built-in battery and a sign that it supports even lower power consumption.  On Sunday, Ukrainian repair expert Oleg Kutkov reported discovering the “mini2_prod1” in the latest firmware release for the Starlink dish hardware. The big upgrade is the addition of the built-in battery, which the current Mini model lacks, requiring customers to source the component from third-party accessory makers. Kutkov’s finding confirmed the new Mini boasts a built-in four-cell lithium-ion battery. PC Mag

Nearly a year ago, Elon Musk-owned X announced it would begin rebuilding the Android version of its app, which had not held up well compared with its iOS counterpart. On Monday, the company shipped the refreshed app, which is now available to download. The new Android version of X was built from scratch and promises improvements to loading, scrolling, notifications, and more, X said in its announcement.

The UK will not have enough water for future datacentres, the water industry has said in stark criticism of the government’s AI growth plans. Datacentres rely on large amounts of water to manage the heat generated by densely packed servers. Cooling towers, chillers and humidification systems make direct use of this water, with large volumes also consumed indirectly through high electrical power needs. Water UK, the trade body that represents water companies in Britain and Northern Ireland, said in a publicly available written briefing to MPs that the government’s water forecasts were “fatally flawed” and “explicitly exclude” datacentres. Guardian 

Almost 3,000 websites have been blocked or seized for illegally streaming World Cup matches, the US Department of Justice (DOJ) has said. More than 1,000 domains were shut down in the US alone during the tournament, with a similar number blocked in Colombia. Enforcement agencies in America and across South America carried out the action under investigations named “operation offsides” and “operation red card”. Ivan J. Arvelo, director of the National Intellectual Property Rights Coordination Center (NIPRCC), said unauthorised broadcasting of World Cup matches violated intellectual property rights and “fuels criminal organizations”. BBC 

Garmin CIRQA was first seen in a photo just a few hours ago, along with indications that the fitness and health wristband will be officially unveiled as early as Tuesday, July 21. Reddit user jneuveglise has now uncovered additional images reportedly sourced directly from an official Garmin website. These images show the Garmin CIRQA for the first time in four colors: beige, blue, red, and black. However, Garmin only varies the color of the fabric and the clasp. Notebook Check 

The post Google unveils Pixel Glow, Starlink ‘Mini 2’ dish in production appeared first on Tech Digest.

Hackers could use cheap Chinese AI to launch devastating cyberattacks

The post Hackers could use ‘cheap Chinese AI’ for cyberattacks, Samsung to announce wider Fold 8 appeared first on Tech Digest.

The European Commission has handed Chinese online retail giant AliExpress a record €550 million (£470 million) fine for failing to prevent the sale of illegal, unsafe, and counterfeit products. 

The penalty is the largest ever issued under the European Union’s Digital Services Act (DSA), surpassing previous fines levied against platforms such as Temu and X.

The European Commission’s exhaustive two-year investigation concluded that AliExpress fell significantly short of its obligations under the DSA to assess and mitigate systemic risks. Regulators found that the platform’s detection systems were deeply flawed, allowing millions of prohibited items, such as counterfeit clothing, dangerous cosmetics and unsafe children’s toys, to circulate freely.

Furthermore, the Commission discovered that AliExpress severely under-resourced its human moderation teams, leaving reviewers with mere seconds to assess complex product listings. Investigators also noted that the platform’s recommendation algorithms actively promoted illicit goods, and items flagged as illegal frequently remained visible to consumers for weeks.

Henna Virkkunen, the EU’s executive vice-president for tech sovereignty, security, and democracy, condemned the failures. “The spread of counterfeit clothing, unsafe toys, dangerous cosmetics and other illegal and harmful products is not an unavoidable cost of shopping online – it is a failure by AliExpress to comply with its obligations,” Virkkunen said. “Scale is not an excuse; risks must be identified and addressed systematically.”

A massive player in the European e-commerce landscape, the platform – which is owned by Alibaba- boasts 193 million monthly active users within the EU. This makes it significantly larger than its major low-cost competitors in the bloc, including Shein, which has 156 million users, and Temu, which serves 130 million.

Because of its colossal market reach, EU regulators argued that structural failures pose a particular threat to consumer safety. While the €550 million penalty is the highest handed down under the DSA to date, it represents less than 1% of Alibaba’s total global revenue, falling well below the maximum possible fine of 6%.

AliExpress immediately criticised the decision, calling the financial penalty “disproportionate” and arguing that it fails to reflect proactive safety measures already implemented. The company is currently reviewing the ruling, but must pay the fine and submit a comprehensive compliance action plan to the European Commission by October 20, 2026.

The post EU fines AliExpress record €550m over selling illegal goods appeared first on Tech Digest.

Key findings:

• More than one in four (27%) UK media consumers say the amount of media they consume harms their wellbeing.
• Spending more time consuming media than intended is the most commonly cited negative effect, reported by 51% of UK adults.
• Younger adults are significantly more likely to associate media consumption with negative outcomes.

More than one in four UK adults believe that the amount of media they consume harms their overall wellbeing, according to a new report from YouGov. 

While nearly half of the population (49%) maintains a neutral stance, and 20% view their consumption positively, concerns regarding screen time and digital habits are increasingly widespread.

The findings form the fourth instalment of a comprehensive media consumption trends series from research company YouGov, highlighting the psychological and behavioural challenges linked to modern screen use.

The most frequently cited issue is unintentional overconsumption, with 51% of adults admitting they spend more time consuming media than originally intended.

Other common negative side effects include sleep disruption (reported by 31%), reduced attention spans and focus (28%), diminished productivity (26%) and information overload (24%).

Mental wellbeing concerns are also prevalent, with 19% of consumers linking their media habits to increased stress or anxiety, and 17% reporting feelings of low mood, depression, or unhealthy social comparison.

The data reveals a stark generational divide, with younger adults experiencing the heaviest psychological toll. Among 18 to 24-year-olds, 46% struggle with sleep disruption, 43% report a reduced attention span and 39% cite lowered productivity due to media habits.

Despite these pressures, this younger demographic is also the most likely to view media consumption positively, with 29% noting an upside, compared to just 18% of adults aged 55 and over.

Gender differences also emerge in the data, particularly regarding digital habits. Women are noticeably more likely than men to exceed their intended screen time (55% compared to 47%) and to experience negative social comparison while browsing platforms (22% versus 12%).

Ultimately, while older demographics largely view their digital consumption with indifference – 57% of those over 55 report a neutral impact – the study underscores mounting anxiety over screen addiction and cognitive fatigue, particularly among the nation’s youngest digital consumers.

https://yougov.com/en-gb/articles/55124-uk-media-consumption-trends-2026-do-brits-think-social-media-and-screen-time-affects-wellbeing

The post 1 in 4 Brits believe media consumption has negative effect on wellbeing appeared first on Tech Digest.

Unlocking a phone with a glance or a fingertip has become so routine that most people barely register it as security anymore. Yet behind that split-second gesture sits a genuine authentication battle: biometric logins versus digital ID verification. Both are expanding fast, but they solve different problems and win consumer trust in different ways.

Understanding the distinction matters more than ever. Passwords remain a weak link, and services across banking, retail, and entertainment are racing to replace or reinforce them with something harder to steal.

Biometric Authentication Gains Ground In Devices

Biometric login relies on something you are rather than something you know. Face scans, fingerprints, and even voice patterns unlock devices and approve payments in under a second, with the underlying data typically staying on the device itself. That local storage model is precisely why biometrics feel low-friction: there is no password to forget and, in most cases, nothing transmitted over a network during routine use.

This convenience has pushed biometric sign-in far beyond phones and laptops. Mobile wallets, banking apps, and entertainment platforms have all moved toward biometric sign-in. Online casino platforms focused on smooth registration and blockchain-based security are part of that same shift like those listed at (source: https://99bitcoins.com/best-bitcoin-casino/no-kyc-casinos/). The pattern is consistent across sectors: once a device supports biometrics, users expect it everywhere.

Digital ID Checks Expand Across Online Services

Digital ID verification takes a slower, more thorough approach. Instead of a quick biometric match, it typically combines document scanning, liveness detection, and database checks to confirm a person’s real-world identity before granting account access. This process usually happens once, during onboarding, rather than at every login.

That thoroughness is why regulators and financial institutions increasingly treat digital ID verification as the gold standard for higher-stakes moments. The market reflects that shift clearly. According to a recent market analysis, the identity verification sector was valued at 11.21 billion dollars in 2023 and is projected to reach over 40 billion dollars by 2032 as businesses invest heavily in stronger onboarding tools. Fraud pressure is a major driver of that spending: nearly 29% of US adults experienced account takeover last year, according to industry fraud research, underscoring why companies are hardening both login and identity-proofing systems simultaneously.

Where Verification Friction Shapes User Behaviour

Friction is the real dividing line between these two technologies. Biometric logins are nearly invisible once set up, which is exactly why users tolerate them for frequent actions like checking a banking app or approving a payment. Ask someone to type a sixteen-character password fifteen times a day and they will look for alternatives fast.

Digital ID verification, by contrast, introduces visible friction at exactly the moments users are most cautious: opening a new account, verifying a large transaction, or responding to a risk flag. That friction can build trust rather than erode it, provided the reason for the extra step is clear. Document verification platforms note that combining document checks, biometric comparison, and liveness detection creates a more defensible identity trail, as outlined in detailed verification guidance, which helps explain why consumers generally accept a slower process when the stakes are higher.

Balancing Convenience And Privacy For Consumers

Neither technology wins outright because they are built for different jobs. Biometrics dominate everyday convenience: fast, local, and nearly frictionless for repeat access. Digital ID verification dominates higher-stakes assurance: slower, more thorough, and better suited to onboarding or high-value transactions where regulators demand stronger proof.

The broader identity infrastructure supporting both is growing quickly. Global digital identity solutions are expected to climb from roughly 47 billion dollars in 2025 to more than 231 billion dollars by 2035, according to projected market growth, as more sectors fold verification into standard consumer journeys. For everyday users, the practical takeaway is straightforward: trust biometrics for speed, trust digital ID checks for security, and expect both to keep merging into a single, increasingly seamless authentication experience.

The post Biometric Logins vs Digital ID Verification: Which Wins Trust appeared first on Tech Digest.

Chinese AI start-up Moonshot

After higher Galaxy S26 prices, and a likely more expensive Galaxy Z Fold 8, it looks like the RAM crisis has claimed another victim: Google’s Pixel 11. The main reason to buy a Google phone: its comparatively cheap price, may soon be lost. Deleted Amazon listings, spotted by an Android Authority reader and corroborated by Droid Life, reveal the base Pixel 11 starting at $899 for 256GB, $100 more than the iPhone 17’s $799 starting price. An Apple phone being cheaper at launch than a Pixel is a collector’s item. Forbes 

Andy Burnham’s plan to scrap the government’s technology department has triggered an angry backlash from MPs, Whitehall officials and tech experts. The incoming prime minister has asked officials to draw up plans to abolish the Department for Science, Innovation and Technology as part of a wider Whitehall shake-up. The plans have not yet been signed off, but they are causing significant disquiet among experts inside and outside government. The Guardian 

“Culture is the only sustainable competitive advantage a business has,” Simon Rogerson, the chief executive and co-founder of Octopus Group, wrote in a LinkedIn post in December last year. “But amazingly, very few businesses have actually worked this out.” The post surprised employees of Vitrifi, a telecoms company, whose 60-odd workers had found out over Zoom just days earlier that they were being made redundant. Telegraph 

Readers may already be familiar with the Garmin CIRQA fitness tracker, even though it has neither been officially announced nor released. Certifications strongly indicate that a market launch is approaching, and there is already reliable information about its feature set. Now, even an image of the wearable has surfaced. As with previous leaks, however, this information should be treated with caution until officially confirmed. The image was discovered by Andre Lardon within the Garmin Connect app. Notebook Check 

In the deserts of the United Arab Emirates a sprawling clean energy project, stretching across an area roughly the size of 12,600 football fields, will play host to a breakthrough allowing solar energy to power the equivalent of half a million homes through the night. The Gulf state has been steadily combining 5.2GW of solar power capacity with 19GWh of battery storage to create the largest battery scheme in the world. Meanwhile, about 7,500 miles away, at the US’s National Renewable Energy Laboratory facility in Colorado, researchers and engineers are making some of the smallest batteries the world has ever seen. The Guardian 

The post China’s Moonshot can rival US firms, Google Pixel 11 prices leaked appeared first on Tech Digest.

Virgin Media has introduced a new fleet of AI-powered “robots” designed to monitor its television network in real-time, aiming to identify and resolve service issues before they impact the viewer experience.

The technology works by simulating the consumer experience, continuously watching and auditing over 220 IPTV channels. By employing automated monitoring, the AI proactively checks for any degradation in video or audio quality.

Should the system detect an irregularity, it immediately triggers an alert to human technical teams, allowing for significantly faster investigation and resolution of faults than traditional monitoring methods, claims Virgin.

Under the new protocol, Virgin Media’s most popular channels are audited by these AI agents every six minutes. This high-frequency check ensures that the company maintains granular visibility into its network performance, providing greater consistency for live sports, season finales, and other high-stakes broadcasts.

Jeanie York, Chief Technology Officer at Virgin Media O2, highlights the strategic value of this innovation. “Creating and launching these new TV robots means we’re able to identify and address any service issues faster and more accurately than ever before, and in some cases before a customer even realises there was a problem,” says  York.

“This new capability is another innovative step in giving our customers a great experience and ensuring we provide the reliable TV service they expect.”

The deployment is part of a broader push by the company to integrate artificial intelligence into its operational framework. Following this initial rollout, the engineering team is already exploring future enhancements for the system.

Potential developments include training the AI to identify a wider, more complex range of service issues and further strengthening the network’s self-healing capabilities. By automating the quality assurance process, Virgin Media aims to boost customer confidence and minimise disruptions across its digital TV landscape.

The post Virgin Media deploys AI robots to enhance TV reliability appeared first on Tech Digest.