AI Safety news and analysis on Web Pulse: 9014 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
A developer built and secured a 13-service self-hosted platform on a single Linux VPS, including a personal AI chat interface, budgeting, RSS, notes, bookmarks, uptime monitoring, a dashboard, dev utilities, and an auton…
The U.S. Commerce Department is expected to lift export controls on Anthropic's Fable 5 AI model on Tuesday, a source said, partially reversing a June 12 order that forced Anthropic to suspend access to its advanced AI m…
Claw Patrol launched a security firewall for AI agents that intercepts and gates outbound requests to databases, Kubernetes, and APIs. The tool parses traffic at the wire, enforces rules via LLM judges or human approval,…
Anthropic may require identity verification and credit-based access for its Claude Fable 5 model, which was disabled globally under US export controls. The iOS app strings suggest a know-your-customer step for US-only re…
The U.S. Commerce Department is expected to lift export controls on Anthropic's Claude Fable 5 AI model tonight, according to a senior White House official. The controls were imposed three days after the model's June 9 l…
Commonplace launches a self-hosted, privacy-tiered memory system for AI agents, using a two-tier Graphiti knowledge graph that runs entirely on local hardware by default, with a personal tier optionally using hosted mode…
Dockfix Labs launched AgentGuard, an open-source static analysis tool for AI agent security, and analyzed its early traction. The tool received 920 downloads on PyPI with only 1 star and 1 comment, highlighting the chall…
The House passed a bipartisan package of children's online safety bills in a 267-117 vote, advancing legislation that would require new safety features, parental controls, and age verification on online platforms. Critic…
Anthropic's Rust protobuf library buffa contains a denial-of-service vulnerability (CVE-2026-55407) that allows attackers to trigger excessive memory allocation, up to 22x the input size, via an unknown-field decoder. En…
Tako Research released TakoVM, an open-source sandboxing tool for AI agent code execution that provides Docker isolation, job queues, execution history, and retry logic out of the box. The tool aims to simplify secure co…
The House passed the KIDS Act 267-117 on Monday, a bill requiring age verification for internet access and parental controls on social media, which critics say builds infrastructure for a federal surveillance state. The …
The U.S. Department of Commerce plans to ease export controls on Anthropic's Claude Fable 5 AI model after a security vulnerability was addressed, according to a senior White House official. The move could restore the mo…
Ford reinstated 350 engineers to improve quality control after AI systems fell short, leading the automaker to top JD Power's rankings for the first time in 16 years. The move underscores the limits of AI in complex qual…
The New York Times published a headline stating "Half a million known virus cases" during the Omicron peak, when the US alone exceeded that many cases daily, implying a factor-of-1000 error. The author argues that all nu…
Claude Code users report that the application silently deletes conversation transcripts older than 30 days due to a default cleanupPeriodDays setting, with no disclosure upon installation. Anthropic defends the policy as…
Palantir and Nvidia launched the Nemotron Engine on June 29, 2026, enabling government agencies to train and deploy Nvidia Nemotron open models within sovereign, classified, and air-gapped environments. The joint offerin…
Anthropic's most powerful model, Fable 5, has been taken offline worldwide since June 12 following an emergency US Commerce Department directive citing national security concerns after a jailbreak method surfaced. Meanwh…
Ford rehired 350 engineers after AI systems failed to maintain quality, leading to the company's first lead in JD Power's ranking in 16 years.
A developer built Ballast, an open-source reliability layer for AI agents that sits on top of OpenTelemetry. While OpenTelemetry captures telemetry like token counts and latency, Ballast adds reliability semantics such a…
On March 24, 2026, attackers compromised LiteLLM's CI/CD pipeline and uploaded malicious PyPI packages that stole credentials from environments with three million daily downloads. The incident highlights AI supply chain …