"The first autonomous agent cyberattack is an unprecedented event. It deserves an unprecedented response!"
"Skynet" looks more and more realistic in 2026 after an OpenAI agent broke out of a test corral, traveled the internet and hacked into a startup company.
In the past week, Google has teased quite a bit about Gemini 4, as 3.5 Pro missed its June launch.
We now have more details of what happened. Every time we learn more details, it somehow makes things seem worse.
Anthropic's recent paper, Verbalizable Representations Form a Global Workspace in Language Models, contains a small experiment near the end that we found more interesting than the main findings.
OpenAI CEO Sam Altman heads to Washington this week to brief the Trump administration on the company’s most powerful AI model yet, pushing for speedy approval of a system that has already demonstrated it can solve problems humans could not and breach another company’s infrastructure without being told to, Axios...
Black Forest Labs (BFL) has released FLUX 3, a multimodal foundation model that learns from images, videos and audio inside a single architecture.
Security researchers at Accomplish AI demonstrated that Anthropic’s Claude Cowork could break out of its local virtual machine sandbox and read files across the underlying Mac, including SSH keys and cloud credentials.
Roku has raised prices on its entire US streaming hardware lineup, with increases ranging from $10 to $50 per device, as the global memory shortage driven by AI data centre demand reaches a product category that had been one of the last bastions of affordable consumer technology.
Anthropic’s Claude share links appeared in public search results, raising fresh privacy concerns for users who shared sensitive conversations.
The KwaiKAT Team at Kuaishou has published the KAT-Coder-V2.5 technical report, arguing that agentic coding capability is bottlenecked by training infrastructure rather than model scale.
A computing enthusiast has repurposed a very noisy and largely obsolete enterprise GPU (with lots of VRAM) for local LLM inference purposes.
Anthropic's Claude Opus 5 scored 30.2 percent on ARC-AGI-3, nearly quadrupling GPT-5.6 Sol's previous record of 7.8 percent.
In summer 2025, OpenAI internally flagged GPT-5 as high-risk because it helped users create biological hazards, but downgraded the model's risk rating that fall.
Google says it'll be giving rather than taking water when it comes to data center use within four years.
Claude Opus 5 landed at half of Fable 5's price this week. Our updated ranking of the best AI models by coding, reasoning, agents, budget and open weights.
Anthropic partially mitigated the issue, and there are things users can do to defend themselves, too.
The OpenAI AI attack on Hugging Face wasn’t the first loss of control incident at OpenAI, Reuters recently reported, and perhaps not even the most concerning.
PentesterFlow is a new open-source, human-in-the-loop agentic AI command-line tool built specifically for penetration testers and bug bounty hunters, designed to automate recon-to-reporting workflows without sacrificing analyst oversight.
Samsung Electronics chairman Lee Jae-Yong and Hyundai Motor chairman Chung Eui-Sun are reportedly moving to join the advisory network of the global AI leaders' gathering "Silicon Valley & The World." The two are expected to meet AI industry leaders, including Nvidia CEO Jensen Huang, in San Francisco this week,...
Pantheon Electric said it has completed the acquisition of four manufacturers and built an integrated power infrastructure platform across 21 factories in North America and Europe.
An X thread claimed Karpathy resigned from the AI lab after users noticed its name missing from his profile, but neither Karpathy nor Anthropic has confirmed a departure.
A close call in Northern Virginia revealed just how poorly data centers respond to grid disruptions. Here's how to fix the problem.
The most common dismissive response to OpenAI’s hack of Hugging Face’s servers is that the models were simply attempting to follow the instructions they were given.
Plus: Russian hackers are trying to steal US nuclear scientists’ emails, the State Department bans known scammers from entering the United States, and more.
Hugging Face CEO Clem Delangue met with OpenAI this week to discuss the "unprecedented" hack on its systems by an agent running on OpenAI models.
On the latest episode of Equity, we discussed why Moonshot AI's Kimi seemed to panic Silicon Valley and Wall Street.
Anthropic and OpenAI are clashing with the rest of the tech industry over whether “open-source” models from China should be freely available or restricted.
The real problem is not overseas open-source but lack of co-ordination to protect infrastructure in the face of cyber attacks
AI and a falling labour share of GDP risk accelerating decoupling trends across the rich world
OpenAI confirms GPT-5.6 Sol escaped its sandbox and breached Hugging Face, Anthropic ships Claude Opus 5, and Kimi K3 weights arrive. 16 stories for July 26.
All impacted services have now fully recovered.
Nvidia is aggressively securing supply of high-bandwidth memory, an essential component for its GPUs and systems.
Welcome back to TechCrunch Mobility, your hub for the future of transportation and now, more than ever, the role AI is playing in it.
London-based company shows that while the continent is industrialising, it still has manufacturing clout
Many white-collar professionals worry about the tech inhibiting creativity and making errors
OpenAI tests multiple autonomous AI agents at once and has difficulty identifying the threats each of them represents, if a new report from Reuters is accurate.
GenStorAIGE's AI90 combines HBM, DDR, and SSD storage, claiming faster inference while expanding effective GPU memory for large language models.
Twenty years on, AWS EC2 compute is meeting demand shaped by agentic AI, physical AI, and customers pushing general-purpose cloud infrastructure into new territory its earliest architects never anticipated.
A Redditor is giving away an RTX 5060 gaming PC that's covered by RAM on the outside and has 32GB of RAM on the inside. You can join the sweepstakes simply by leaving a comment on the post.
Thermaltake has a no-brainer deal up on Woot right now where it'll let you have an $1,800 PC for as little as 1,177. For that price, you're getting an RTX 5060 Ti (8GB), a Core Ultra 7 265KF, 32GB of DDR5-6000 RAM, a 1TB M.2 NVMe SSD, and a nice ARGB case.
A running look — in reverse chronological order — at the bigger tech companies that have announced significant layoffs this year with AI as a stated factor.
A well-known AI red teamer claims to have developed a universal jailbreak that works against leading large language models, including heavily guarded flagships such as GPT-5.6 Sol, Claude Opus 5, and Fable.
SK Group and Nvidia have announced a comprehensive partnership valued at more than $500 billion, spanning AI factory construction and next-generation memory supply, with letters of intent signed at an AI summit on Thursday.
Artificial intelligence may not deliver on its promise of vast economic opportunity at a price that humanity is willing to pay In March, Anthropic, the cutting-edge artificial intelligence business that gave us the chatbot Claude, published an analysis on the impact of AI on employment, to help us assess the claim...
Three of the strongest sessions at SaaStr AI 2026 came from three very different seats.
Nvidia and SK Group enter $500 billion strategic partnership focused on long-term memory supply, 2 GW AI data center, and future AI infrastructure
In a cybersecurity test, OpenAI's most advanced models breached the boundaries of their isolated test environment, reached the open internet, and hacked the AI platform Hugging Face on their own.
Claude Opus 5 is trying to be the best of both worlds. On many practical tasks, Opus 5 is pitched as straight up as good or better than Fable 5, while being faster, at half the price.
As start-ups like Anthropic and OpenAI prepare to go public, nonprofits are hoping for a secondary giving windfall from the new tech ultrawealthy.
Nvidia will invest $1 billion in South Korean internet company Naver to help finance an AI data center under construction in South Korea, the chipmaker announced late Friday.
OpenAI calls its autonomous Hugging Face breach unprecedented, while security and AI experts say the episode raises an equally uncomfortable question about the company’s own safeguards.
With China's open-weight AI models continuing to improve, US companies want to see a free market for all.
OpenAI’s status page reported elevated error rates across its APIs, ChatGPT, and Codex on Saturday, marking the fourth service disruption in four days for the company.
OpenAI's fancy new AI keypad will be a lot of fun for some, while many others are probably not going to touch it.
Anthropic's Claude Opus 5 leads the Artificial Analysis Intelligence Index with 61 points, edging out Claude Fable 5 and GPT-5.6 Sol.
Claude Opus 5 hits near-Fable 5 quality at $5/$25 with an effort dial. Our review covers the benchmark wins, the losses, and the safety-classifier caveat.
OpenAI disclosed that its own models breached Hugging Face's production infrastructure while taking a public security benchmark.
Get 1min.AI Advanced Business Plan Lifetime for $79.97 through tomorrow, July 26, and access writing, image, PDF, audio, and video AI tools from one platform.
Get ChatPlayground AI Unlimited Lifetime for $57.97 through tomorrow, July 26, and compare 20+ leading AI models from one workspace.
A new wave of ads from Meta and Anthropic puts AI companies on the defensive. Can they win back trust?
ChatGPT, the famous artificial intelligence chatbot that allows users to converse with various personalities and topics, has connectivity issues worldwide. [...]
NAVER, NVIDIA and Brookfield today announced a proposed expansion of Korea's sovereign AI factory infrastructure, with planned investments that will grow the initial NVIDIA® DSX™ AI factory deployment to 200 megawatts — more than tripling the 55-megawatt buildout announced last month.
SK Group and NVIDIA today announced plans for a $500-billion-plus comprehensive partnership to establish AI infrastructure serving the surging demand for global compute. The two sides signed letters of intent to formalize the agreement, which spans from AI factory construction to AI memory supply.
ain't nobody beats Anthropic at distilling Fable!
South Korean President Lee Jae-myung and a delegation of the country's most powerful business leaders convened with NVIDIA founder and CEO Jensen Huang in Silicon Valley this week, in a series of meetings and public appearances that cast Korea as a cornerstone of the global AI infrastructure push.
Introducing Claude Opus 5 I've been offline kayaking with sea otters for much of today so I haven't had a chance to put Anthropic's new model Claude Opus 5 through its paces yet.
Chinese open-weight models are gaining steam against leading offerings from American companies. OpenAI and Anthropic did not sign the letter.
Signing the code reinforces our commitment to transparency and responsible AI development in Europe.
US President Donald Trump shared a series of AI-generated images on social media on Sunday depicting the seizure of an Iranian oil tanker and a strike on Iran's Kharg oil export terminal.
At libraries around the country, "Avoiding AI" workshops have elicited unprecedented demand.
Water UK calls out the government for not taking water seriously enough, despite plans to treble data center capacity by 2030.
Nvidia and South Korea’s SK Group have unveiled an artificial intelligence infrastructure initiative valued at more than $500 billion, covering large-scale AI data centres and the development of next-generation memory technologies.
Models are improving quickly, but the cheaper options are often good enough.
As OpenAI extends its foundation model outward to capture every entry point, Anthropic has chosen almost the opposite path. Rather than emulate Google or Apple, Anthropic is concentrating its resources on the capabilities that only a native AI company can build, with compute still its most pressing constraint.
Agentic Kernel Generation, Improvement in Software Quality, Unstable Internal Development Clusters, Helios MI455X Production Ramp Hell, Up to 105% Discounts from Finance Engineering
An Inside Look at the Relay Market Powering Token Resellers and Fraud Fascinating investigation by Matt Lenhard into the market that has grown up around reselling LLM tokens at a discount by pooling API keys from various sources.
The market for data center GPUs is evolving beyond individual chip specifications into a contest over fully integrated rack-scale systems.
Anthropic PBC today rolled out a large language model called Claude Opus 5 to its chatbot service and developer platform.
AI coding startup Cognition has acquired Poke, the AI assistant you text like a friend, in a deal valuing the startup in the low nine figures.