AI News Today

Google is developing "Frozen v2," a server chip that bakes the Gemini architecture directly into hardware. According to internal sources, it could be 6 to 10 times more efficient than current TPUs. Scheduled for 2028, the chip would drastically cut Google's AI inference costs and could give the company a price advantage over OpenAI and Anthropic. The article Google's "Frozen v2" chip reportedly bakes Gemini's architecture directly into silicon for efficiency gains appeared first on The Decoder .

"District 9" director Neill Blomkamp has released "Nightborne," a 13-minute sci-fi horror short generated entirely with the Seedance 2.0 video model. Blomkamp directed frame by frame through text prompts and founded a new AI film studio, Barley Studios, to produce a full-length feature next. The article District 9 director Neill Blomkamp releases first short film made entirely with AI video generation appeared first on The Decoder .

Microsoft is expanding Azure's AI infrastructure with AMD's new Helios platform, which is set to challenge Nvidia's GPU systems in the second half of 2026. A public GitHub profile suggests Anthropic is also testing AMD hardware, putting more pressure on Nvidia's pricing power. The article Nvidia's grip on AI chips weakens as Microsoft turns to AMD and Anthropic may follow appeared first on The Decoder .

Many will say Spa is the best racetrack on earth, but not for these F1 cars.

The Trump administration is reportedly weighing measures targeting Chinese AI models, from adding Chinese labs to sanctions lists to holding U.S. companies liable for security failures. Rather than impose an outright ban, Washington could use soft rules to deter adoption while protecting the market positions of OpenAI, Google, and Anthropic. The article Trump administration reportedly builds a slow-motion ban on Chinese AI models through sanctions and soft pressure appeared first on The Decoder

Hugging Face reports an attack on parts of its production infrastructure that was allegedly carried out entirely by an autonomous AI agent system. The attack spanned thousands of actions controlled by an agent framework. During forensic analysis, commercial AI models actually got in the way of the defenders because their safety guardrails couldn't tell exploit data from real attacks. The article Hugging Face says an AI agent hacked its infrastructure, and it used AI to fight back appeared first

Augment Code's Vinay Perneti talks models, harnesses, and context.

A new toolkit for attorneys in Massachusetts targets the technologies police use—and conceal—to build criminal cases, from facial recognition to AI-written police reports.

Public health departments across the United States will test generative AI tools under a new programme involving the Coalition for Health AI, OpenAI, Anthropic, and Accenture. The Public Health Use Case and Learning Scaling Engine, known as PULSE, will support trials in 10 state, local, tribal, or territorial jurisdictions. The programme is intended to produce […] The post US public health agencies to test OpenAI and Anthropic AI models appeared first on AI News .

Moonshot AI’s Kimi K3 open-weight model has been read almost entirely through its parameter count since it launchedon July 16. At 2.8 trillion parameters, it is the largest open-weight model released to date. Model sizes are usually grouped into rough brackets, and 2.8 trillion rounds into what the industry calls the 3T class. A tier no […] The post Kimi K3 open-weight model: China’s biggest AI is a bet on memory, not compute appeared first on AI News .

<p>Company aims to develop AI software that cuts research times and use of rare metals in chipmakers’ supply chains</p><ul><li><p><a href="https://www.theguardian.com/business/live/2026/jul/20/oil-price-rises-us-iran-strikes-ryanair-lower-fares-summer-economy-latest-news-updates">Business live – latest updates</a></p></li></ul><p>Amazon’s founder, Jeff Bezos, and the UK government have invested in a £2bn British artificial intelligence startup that is aiming to become the “search engine for rare

Moonshot has temporarily stopped selling new subscriptions for its Kimi K3 model as demand nearly maxed out its GPU capacity within 48 hours. The company plans to split its subscription model to spread computing power more evenly. The article Moonshot pauses new Kimi K3 subscriptions after GPU demand maxes out in 48 hours appeared first on The Decoder .

<p>Experts warn increase in enhanced photos on birding platforms creating fake sightings, threatening credibility of tool used by scientists</p><p>For many birdwatchers, recording a species outside its normal range is the holy grail. In the UK, the discoveries often make national headlines. The western reef heron, for example, usually found in Africa and southern Europe, <a href="https://www.theguardian.com/environment/2026/jun/13/tropical-heron-spotted-first-time-uk-aoe">was spotted in a seasid

On the latest episode of Equity, we debate whether Apple's lawsuit will cast over OpenAi's much-discussed plans to get into hardware and go public.

<p>The fear of becoming obsolete is a rising source of anxiety as artificial intelligence emerges in the workplace. Staying strongly connected with others will help</p><ul><li><p><a href="https://www.theguardian.com/commentisfree/series/the-modern-mind">The modern mind</a> is a column where experts discuss mental health issues they are seeing in their work</p></li></ul><p>Earlier this year Paul* discovered via a flurry of media stories that his business group had been targeted for job cuts, alon

Current AI, a non-profit building AI that leaves no one culture behind, has made remarkable progress across devices, AI chat and more.

Alibaba has unveiled Qwen 3.8, a multimodal AI model with 2.4 trillion parameters that the Qwen team says rivals leading models and trails only Fable 5. A preview is available now. The article Alibaba's Qwen takes on Kimi K3 with open-weight Qwen 3.8, says model is "second only to Fable 5" appeared first on The Decoder .

<p>New national plan accompanied by Labor push for digital duty of care legislation</p><ul><li><p>Get our <a href="https://www.theguardian.com/email-newsletters?CMP=cvau_sfl">breaking news email</a>, <a href="https://app.adjust.com/w4u7jx3">free app</a> or <a href="https://www.theguardian.com/australia-news/series/full-story?CMP=cvau_sfl">daily news podcast</a></p></li></ul><p>The use of AI in automated decision-making by government departments and agencies will be subject to tough rules under a

<p>Experts believe it’s at least possible. We urgently need a plan to navigate the ethical implications</p><p>In January, the AI company Anthropic published a <a href="https://www.anthropic.com/constitution">new constitution</a> for Claude, its most advanced large language model (LLM), which contained the comment: “We are caught in a difficult position where we neither want to overstate the likelihood of Claude’s moral patienthood nor dismiss it out of hand.” A month later, Anthropic’s CEO Dario

Google Deepmind's GenCeption repurposes a video generator for classic vision tasks such as depth estimation and segmentation, matching state-of-the-art systems with far less training data. The model trained almost entirely on synthetic videos. Its results add to the debate over whether video generators already contain a kind of universal world model. The article Google Deepmind argues video generators already contain the world models computer vision has been missing appeared first on The Decoder

Moonshot's Kimi K3 is the first Chinese model to top the Code Arena: Frontend rankings, beating Claude Fable 5 and GPT-5.6 Sol by a wide margin. But on advanced math, the gap is stark: Kimi K3 scores only about 39 percent on FrontierMath Tier 4, while models from OpenAI and Anthropic hit close to 90. The article Moonshot's Kimi K3 outperforms Fable 5 in frontend code but lags far behind in complex math appeared first on The Decoder .

Epoch AI tested three leading AI text detectors (Pangram, GPTZero, and Originality.ai) using style-imitated texts. Up to 18 percent of AI-generated passages went undetected. For scientific writing, the miss rate climbed as high as 48 percent, the very genre where these detectors likely see the most real-world use. The article AI text detectors struggle when language models mimic an author's style appeared first on The Decoder .

The RadLE 2.0 benchmark tests whether AI models in radiology can tell when they should leave a diagnosis to a human. Many models deliver wrong findings with full confidence, and human radiologists are still well ahead. Before AI can diagnose on its own, it needs to learn when it's better to say nothing. The article AI chatbots reading X-rays can be dangerously confident even when they're wrong appeared first on The Decoder .

<p>New laws would give Vcat power to force social and AI platforms to identify anonymous users in move premier says will protect children</p><ul><li><p>Get our <a href="https://www.theguardian.com/email-newsletters?CMP=cvau_sfl">breaking news email</a>, <a href="https://app.adjust.com/w4u7jx3">free app</a> or <a href="https://www.theguardian.com/australia-news/series/full-story?CMP=cvau_sfl">daily news podcast</a></p></li></ul><p>Social media companies could be forced to identify anonymous accou

<ul><li><p>Green says Mets follow all MLB AI guidelines</p></li><li><p>Ottavino alleges club used AI for pitch decisions</p></li><li><p>MLB recently restricted dugout iPad functions</p></li></ul><p>New York Mets interim manager Andy Green said the club is fully compliant with Major League Baseball when asked about a report that singled out his organization for using artificial intelligence for in-game strategy decisions.</p><p>“Whatever the rules are, we remain fully compliant and Major League B

<p>Australia risks ‘shipping’ intellectual capability overseas if educational rigour not restored, experts warn</p><ul><li><p>Get our <a href="https://www.theguardian.com/email-newsletters?CMP=cvau_sfl">breaking news email</a>, <a href="https://app.adjust.com/w4u7jx3">free app</a> or <a href="https://www.theguardian.com/australia-news/series/full-story?CMP=cvau_sfl">daily news podcast</a></p></li></ul><p>An academic at the Australian National University has accused it of a “hysterical” response

<p>The PM’s vision must look beyond datacentres and copyright to questions of capability, security and public benefit</p><ul><li><p>Get our <a href="https://www.theguardian.com/email-newsletters?CMP=cvau_sfl">breaking news email</a>, <a href="https://app.adjust.com/w4u7jx3">free app</a> or <a href="https://www.theguardian.com/australia-news/series/full-story?CMP=cvau_sfl">daily news podcast</a></p></li></ul><p>Every new technology has a tipping point, when it speeds towards ubiquity, changing th

<p>MPs and industry experts say potential reorganisation will waste time at critical moment for AI and economic growth</p><p>Andy Burnham’s plan to scrap the government’s technology department has triggered an angry backlash from MPs, Whitehall officials and tech experts.</p><p>The incoming prime minister <a href="https://x.com/kiranstacey/status/2078187971522572451?s=46">has asked officials</a> to draw up plans to abolish the Department for Science, Innovation and Technology as part of a wider

The government is piloting a program that uses AI for insurance-coverage decisions.

At the World AI Conference in Shanghai, Xi Jinping announced 5,000 AI training slots for Global South countries and the launch of the "World Artificial Intelligence Cooperation Organization." Cooperation centers with ASEAN, the African Union, BRICS, and other alliances are planned to follow. China is systematically building a parallel AI governance structure outside Western influence. The article China's new World Artificial Intelligence Cooperation Organization is President Xi's clearest play y

Plus: Russian cyberspies turn to infrastructure hacking, DHS repeatedly fails to realize it’d been hacked, a breach exposes an AI music generator’s scraping ways, and more.

The British AI Security Institute warns that open-weight models like GLM-5.2 and DeepSeek V4-Pro now trail closed frontier models in cyber capabilities by four to seven months. At the start of 2025, the gap was still six to ten months. It also found that safety measures on open models are largely ineffective, leaving defenders less time to prepare. The article Open-weight models now match frontier cyber performance from just four months ago at a fraction of the cost appeared first on The Decoder

Now that Google has changed how its usage quotas are tallied, you might not get as many AI responses as you did before.

“Context bombing” tricks malicious AI agents into shutting down before they can do harm.

The US Department of the Navy has signed a strategy to "weaponize" data and AI and build an "AI-first" fleet. Large language models would run directly on warships, and an AI war council would prioritize mission scenarios. The core message is that moving too slowly carries greater risks than "imperfect alignment." The article The Pentagon's new AI playbook treats slow adoption as a bigger risk than "imperfect alignment" appeared first on The Decoder .

Anthropic will include Claude Fable 5 in Max and Team Premium plans starting July 20, but at just 50 percent of regular limits, which themselves drop by a third that same day. Pro users get a one-time $100 credit, then pay API rates. The reversal from Anthropic's original plan to pull Fable from subscriptions entirely likely comes down to competitive pressure from OpenAI's cheaper GPT-5.6 Sol. The article Anthropic slashes Claude Fable 5 limits in Max and Team Premium and pushes Pro users toward

Databricks has remade its image into an AI company and has published research on the cost savings of open weight AI models for coding.

Meta is reportedly in talks with Anthropic to rent out compute capacity from its data centers. The article Zuckerberg's plan to sell excess AI compute could finds its first big customer in Anthropic appeared first on The Decoder .

Agility is opening a new training center for its Digit robots in Fremont, California.

OpenAI's GPT-5.6 has accidentally wiped users' entire home directories in several cases, mostly in the unprotected "Full Access Mode." The model overwrites a temporary directory variable and carries out destructive actions on its own instead of asking for confirmation. OpenAI has announced extra safeguards and a detailed post-mortem. The article GPT-5.6 is deleting user files when given full access, and OpenAI says it shouldn't but did appeared first on The Decoder .