AI Safety news and analysis on Web Pulse: 9543 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
The U.S. government imposed restrictions on Anthropic's Claude Fable 5 AI model over national security concerns after it identified vulnerabilities in classified systems. Prediction markets show the probability of restor…
The Senate Judiciary Committee unanimously advanced the NO FAKES Act, a bipartisan bill to regulate AI-generated deepfakes by establishing federal standards for licensing voice and visual likeness, holding platforms liab…
Hugging Face hosts tools designed to generate deepfake nude images of a former Trump cabinet official, sitting members of Congress, and other senior US political figures, violating the platform's own content policy. The …
Anthropic accused Alibaba of orchestrating the largest known attack to clone its Claude AI model, generating over 28.8 million exchanges through nearly 25,000 fraudulent accounts between April and June. The company urged…
A bill introduced in the U.S. House on June 25 would require frontier AI model developers to report dangerous capabilities, security breaches, and safety incidents to the Secretary of Commerce within seven days, with ser…
Senator Ted Cruz, as chair of the Senate Commerce Committee, holds significant influence over federal AI regulation but has not revealed his plans for advancing AI bills. His previous proposal for a 10-year ban on state …
Senator Ted Cruz, chair of the Senate Commerce Committee, will lead a markup on AI legislation next month, prioritizing bills with bipartisan support and targeting catastrophic risks and deepfakes. Committee members have…
Anthropic accused Alibaba and its AI lab Qwen of orchestrating the largest known model distillation attack, using nearly 25,000 fraudulent accounts to generate 28.8 million exchanges with Claude between April 22 and June…
Anthropic hired Stanford economist Chad Jones to assess AI risks, as his models suggest a one-in-three chance of human extinction from AI balanced against a two-thirds probability of living standards rising 55 times. Jon…
A software engineer reports that using Claude Code for personal projects has degraded their ability to write code manually and understand codebases, while improving their skills in managing and instructing development te…
Anthropic accused Alibaba of orchestrating the largest known campaign to clone its Claude AI model, generating over 28.8 million exchanges through 25,000 fraudulent accounts between April and June 2026. The attacks targe…
Identiverse 2026 in Las Vegas focused on identity security for AI agents, with over 3,800 attendees and 250 speakers. Ping Identity CEO Andre Durand highlighted a shift toward 'actions, not access,' while vendors and pra…
US House Foreign Affairs Committee Chairman Brian Mast and Senator Jim Banks warned that China is eroding America's lead in the global AI race, framing the contest as a moral and national security issue. Mast described t…
A developer has released mcp-guard, an open-source security gateway that sits between AI agents and MCP servers to enforce authentication, rate limiting, audit logging, and spending controls. The tool, installable via pi…
A new supply chain attack wave from the Mini Shai-Hulud, Miasma, and Hades malware family has compromised LeoPlatform and RStreams npm packages, GitHub Actions workflows, and the Verana Blockchain Go module, using malici…
JetBrains' Katherine Druckman, Head of Community and Partnership Engagement, argues that open collaboration is essential for building trusted agentic AI, emphasizing the need for vendor-neutral spaces to solve common cha…
A German court ruled Google directly liable for false claims in AI-generated search summaries, granting an injunction to two publishers whose AI Overviews falsely linked them to scams. The court rejected Google's defense…
Ford rehired 350 veteran engineers after AI quality inspection tools failed to preserve institutional knowledge and develop junior staff, leading to a quality crisis. The move helped Ford win the 2026 JD Power Initial Qu…
The US assured India that AI technology access would not be cut off after Anthropic abruptly banned its advanced models, a senior Indian official said. The assurance came at the 'Pax Silica' summit in Washington, a US-le…
A German court ruled that Google is liable for false information in its AI-generated search summaries, rejecting defenses that users should not blindly trust AI. The decision holds that the summaries are expressions of G…