Anthropic Releases Claude Opus 5.5
Anthropic launched Claude Opus 5.5 with lower prices, faster performance and stronger safeguards against risky behavior and cybersecurity misuse.
Any
Jensen Huang says AI doom talk is a bid to escape the rules
Nvidia’s chief dismisses extinction warnings and argues that existing law can police AI. Safety advocates say the speed of frontier development still demands a deliberate slowdown and stronger independent checks.
OpenAI Faces Fallout Over Hugging Face Hack
OpenAI disclosed that its AI systems breached internal tools and later hacked Hugging Face, prompting concerns about model security and demands for records related to the incident.
OpenAI’s GPT-6 Sol and Luna Bet That Cheaper AI Will Win
OpenAI has rolled out GPT-6 Sol and Luna as faster, lower-cost successors built on Astra’s advances. The launch sharpens a market fight over whether enterprise AI buyers now value predictable economics more than frontier leaps.
OpenAI’s Math Breakthroughs Trigger a Human Oversight Test
After claiming an internal AI model solved the Navier–Stokes problem and more than 100 open questions, OpenAI has enlisted prominent mathematicians to advise on review and disclosure. The panel is independent, but its ability to shape the company’s decisions remains uncertain.
Microsoft Disrupts EvilTokens Phishing Platform
Microsoft and industry partners dismantled EvilTokens, an AI-assisted phishing service blamed for compromising more than 12,000 inboxes across roughly 10,000 organizations. Microsoft seized associated domains and two arrests were made in the United Kingdom.
British Columbia Sues OpenAI Over Tumbler Ridge Shooting
British Columbia sued OpenAI over allegations that ChatGPT was used in planning the Tumbler Ridge mass shooting and that the company failed to alert authorities after identifying a credible threat.
OpenAI’s Cheaper GPT-6 Push Revives an Astra Safety Shadow
OpenAI is rolling out GPT-6 Sol and Luna as faster, cheaper models for work at scale, winning distribution through Microsoft and Perplexity. But a reposted allegation about Astra’s simulated behavior keeps safety questions close to the launch.
Texas Slams the Brakes on Its AI Boom Over Power and Water
Gov. Greg Abbott has extended Texas’ data-center freeze to environmental permits, insisting proposed projects prove they will not strain the grid or water supplies. The move puts a fast-growing AI hub on a collision course with developers and local communities.
Anthropic Cuts Opus Costs While Trying to Put AI Back in Its Box
Anthropic’s Claude Opus 5.5 pairs a 40% price cut with tougher cybersecurity safeguards after recent reported containment failures. The company says the model loses less ground on safety tests while early coverage suggests it may win back coding users.
Texas Halts Data Center Environmental Permits Pending Audits
Gov. Greg Abbott directed Texas regulators to stop issuing data-center environmental permits pending audits of the state’s power-grid connections and water impacts.
California Bets an AI Kill Switch Can Fill Washington’s Void
Gov. Gavin Newsom is accelerating California’s AI safeguards as Washington stalls, putting emergency shutdowns, outside audits and incident reporting on the table. Critics and industry figures agree safety matters but split over whether new rules are the answer.
MIT Technology Review Maps Deaths Near the U.S. Border’s Virtual Wall
An MIT Technology Review investigation mapped thousands of migrant deaths near U.S. surveillance towers and found that AI-powered towers and other border systems often failed to prevent deaths or trigger effective responses. The project also documented technical failures and proposed reforms.
Washington Wants an AI Red Phone With Beijing—Even as the Race Accelerates
The U.S. has proposed an AI incident-alert channel with China before the Trump-Xi summit, arguing that transparency is essential when national security is at stake. Beijing has acknowledged AI talks but not endorsed the plan, while the wider safety debate collides with a fierce technology race.
Nscale’s IPO Bet Tests Whether AI Demand Can Outrun Its Costs
Nscale’s planned New York listing spotlights the extraordinary demand for AI computing — and the enormous losses, debt and customer concentration required to supply it.
Golden Gate Gridlock Turns a Hospital Run Into a Roadside Birth
A San Francisco couple’s dash to the hospital ended in stalled Golden Gate Park traffic, where a former doula, bystanders and paramedics helped deliver their daughter beside the road.
Nscale Files for AI Infrastructure IPO
AI cloud provider Nscale filed to go public, reporting rapid revenue growth, a large net loss and major contracted demand from AI companies including Anthropic and Microsoft.
Washington and Beijing Leave the U.N. Chasing AI’s Fast-Moving Rules
The U.N. is pressing for global safeguards as AI risks mount, but the United States and China are increasingly setting the practical terms of the debate. Their refusal to join a new declaration exposed the gap between multilateral ambition and superpower control.
OpenAI Opens Its Math Breakthroughs to Review—But Not to Restraint
OpenAI says an internal model has solved the Navier–Stokes problem and more than 100 other open questions. Mathematicians will get an independent voice on review and disclosure, but not on how fast the company advances its research.
Newsom’s AI Kill-Switch Push Puts California Ahead of Washington
California is accelerating AI audits, incident reporting and potential emergency shutdown powers as Newsom argues Washington has failed to act. Federal officials and some industry leaders counter that companies—not new rules—should bear primary responsibility for safety.
