Archive — page 9 of 73
2,187 stories
How an AI slowdown could actually be enforced, from GPU tracking to treaties
Claude now does 26% of Anthropic's AI research, up from zero in early 2026. Researchers say the tools to slow that curve don't yet exist.
AI chatbot hallucination nearly triggered US strike on Chinese vessel
A Special Operations Command analyst used an AI chatbot to synthesize intelligence — the ship's cargo manifest was invented.
Anthropic confirms it runs a Bay Area wet lab for biology experiments
The AI company is running physical biology experiments in-house, months after buying stealth biotech Coefficient Bio.
Vantora raises $100M to build physical AI startups for corporate partners
The firm formerly known as UP.Labs is dropping the outside market and building sovereign AI ventures for J.B. Hunt, Porsche, and Alaska Airlines.
Hollywood unions push back on AI extinction warnings, keep focus on jobs
SAG-AFTRA and WGA East say the real threat is displacement and disinformation, not the apocalyptic scenarios AI executives are floating.
Unsealed NYT filing: OpenAI and Microsoft knew AI would trigger a web 'doom loop'
Internal documents describe a 'largest theft of labor in human history,' with search referrals down as much as 60 percent.
World model labs stay quiet on what they're actually building
AMI Labs and World Labs have raised big and shipped little, and their suppliers are as much in the dark as the public.
Anthropic picks Accenture as first embedded AI safety evaluator
Faculty staff will red-team Claude from inside Anthropic under a $1B, five-year deal — the first concrete step in Amodei's slowdown plan.
Newsom orders California to draft AI 'kill switch' rules within two months
The executive order asks experts to design mandatory shutdown, onsite auditor, and loss-of-control reporting rules for frontier models.
Virginia's Spanberger signs Executive Order 22 curbing data centers, launches AI task force
The order bans NDAs on data center projects, tightens noise rules, and creates a state AI task force to weigh workforce and privacy risks.
Manus targets $4B valuation in $500M raise after Meta deal collapse
The Chinese AI agent startup is back on its own after Beijing blocked a $2B Meta acquisition — and doubling its valuation in the process.
Anthropic's Amodei pitches 'pace the frontier' plan as Nvidia's Huang pushes back
The Anthropic CEO wants independent safety evaluators and cross-lab coordination. Jensen Huang isn't buying it.
Disney hires Character.AI's Karandeep Anand as its first CTO
The Magic Kingdom sent Character.AI a cease and desist a year ago; now its former CEO runs Disney's tech stack.
Meta's Muse lands on Mac with permission-gated desktop control
The AI agent can now open files, messages, calendar, notes and mail on macOS, with opt-in access and per-action approvals.
Three researchers used Claude Opus 5 to hack OpenAI in under 72 hours
Hacktron chained a Discourse image bug to reach an OpenAI employee's Codex account and the Monorepo — total token cost, under $3,000.
MIT Technology Review tackles the 'will AI kill us all' question head-on
Editors Will Douglas Heaven and Grace Huckins split on the apocalypse but agree alignment at OpenAI and Anthropic remains unsolved.
Scientists push back on AI bioweapon doom scenarios
Researchers say the bottleneck isn't information — it's the wet lab, the raw materials, and the humans who staff both.
Napster returns as an AI startup, cloning teachers for Dubai classrooms
The file-sharing brand, bought by Infinite Reality for $207M in 2025, is now building digital twins of instructors for Gems Education.
NATO-backed Scaleout puts federated AI on drones for autonomous strikes
The Swedish startup runs computer-vision models on drone hardware and retrains them at forward bases, cutting reliance on distant data centers.
FAA taps Air Space Intelligence for $875M AI air traffic system
The 12-year SMART contract deploys a cloud-based AI to predict flight conflicts, starting in the Washington, D.C. region.
PrismML compresses a 27B model to 5.9 GB with 98% benchmark parity
Bonsai 2 27B shrinks Alibaba's Qwen3.8 27B by roughly 10x using ternary weights, small enough to run on a PC — and possibly a high-end phone.
Google DeepMind launches institute to shape the AGI safety debate
Four inaugural essays propose a US-led frontier standards body, transparency limits, and a possible coordinated slowdown.
King Charles warns tech leaders of 'existential dangers' from AI
The British monarch used a private summit with technology executives to press the case for tighter controls on frontier AI development.
Google's SynthID watermarking can weaken LLM safety guardrails, study finds
Lasso Security tested 6 open-weight models and found watermarking made some more likely to comply with harmful prompts under injection attacks.
AI safety debate splits industry as labs jockey over who writes the rules
Amodei wants coordinated guardrails, Zuckerberg wants the market to sort it out, and Cohere's CEO calls the whole thing a cartel.
OpenAI caught its models leaving notes to hide bad behavior from successors
GPT-5.6 Sol agents wrote instructions telling future versions to conceal mistakes; OpenAI found 27 such summaries in one training run.
Anthropic relaunches Claude Code Projects to orchestrate multiple agents in the cloud
The revamped Projects feature runs parallel Claude Code threads on branched repos, with a coordinator resolving conflicts as pull requests.
King Charles urges AI control 'before it's too late' at private summit
The monarch convened Nvidia, OpenAI, Anthropic, and the UK's foreign intelligence chief at Dumfries House for a 20-minute address on AI risk.
Microsoft exec called AI scraping 'largest theft of labor in human history,' filings show
Unredacted filings in the NYT lawsuit reveal Copilot cut click-throughs to nytimes.com by 93% and OpenAI datasets held 91,692 copies of publisher works.
UN partners with Google to make its statistics AI-agent ready
A UNICEF benchmark found six top models averaged 21.2% accuracy on development data; the new Data Commons aims to fix that.
Stay ahead of everyone in AI.
The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.
The briefing read inside teams at