AI Safety news and analysis on Web Pulse: 12886 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
The Pentagon disclosed in a sworn filing that xAI's Grok chatbot was used to support over 2,000 missile strikes within 96 hours, marking a significant deployment of commercial AI in military operations. The filing, repor…
The Commerce Department imposed export controls on two of Anthropic's advanced AI models on June 13, forcing the company to globally disable its Fable 5 and Mythos 5 models. The move reignited congressional debate over A…
France's domestic intelligence service will replace US tech company Palantir's AI data tools with those from French firm ChapsVision to avoid strategic dependency, Prime Minister Sébastien Lecornu announced. The move ref…
CaML, a nonprofit focused on AI alignment, launched community polls to gauge opinions on controversial alignment questions, including whether robust alignment requires pretraining intervention and whether multipolar worl…
The US has delayed blacklisting Chinese AI developer DeepSeek, memory chipmaker CXMT, and over 100 other firms to avoid escalating tensions with Beijing amid ongoing trade negotiations. The Commerce Department's Entity L…
1Password has expanded its MCP Server to Kiro, a software development agent, enabling AI-powered development with runtime-issued, task-scoped credentials that never enter the model's context window. The integration allow…
AI coding tools are reshaping software engineering, shifting focus from writing code to evaluating what to build and judging output quality. Anthropic's Mythos model found thousands of bugs but patched few, highlighting …
Four days after Washington cut foreign access to Anthropic's top models, Cohere reports a surge in government inquiries, DeepSeek closed a record $7.4 billion funding round, and Chinese AI labs slashed token prices up to…
SpaceX agreed to acquire AI coding startup Cursor for $60 billion in stock on June 16, 2026, raising concerns about using the same AI stack for both code generation and review. Industry data shows a 14x increase in AI-ge…
The US Commerce Department ordered Anthropic to disable its Fable 5 and Mythos 5 AI models for all foreign nationals, citing national security concerns over a jailbreak that could weaponize the models' vulnerability-hunt…
The Pentagon disclosed in a Justice Department filing that xAI's Grok chatbot was used to target over 2,000 munitions at 2,000 distinct targets in Iran within 96 hours, calling continued operation of xAI data centers a m…
Databricks acquired cybersecurity startup Panther Labs, which had a $1.4 billion valuation after a 2021 funding round, to bolster its security offerings. The acquisition aims to enhance Databricks' competitive position a…
A comprehensive review published on arXiv examines fine-tuning techniques for Large Language Models (LLMs), covering methodologies from supervised and unsupervised learning to parameter-efficient methods like LoRA and ad…
Executives and technical leaders from the US and its allies urge Commerce Secretary Lutnick and National Cyber Director Cairncross to lift export controls on Anthropic's Fable and Mythos AI models, arguing the restrictio…
The Trump administration ordered Anthropic to suspend access to its Claude Fable 5 AI model for non-U.S. nationals after Amazon reported jailbreaking vulnerabilities to the Commerce Department. The administration rejecte…
Sparda launches a tool that converts Express and FastAPI applications into MCP servers in three minutes without requiring OpenAPI specs, accounts, or hosting. The tool uses AST scanning to inject a reversible MCP router,…
The U.S. Department of Justice on June 16 asked a federal court to dismiss an NAACP lawsuit against xAI's Mississippi data center, arguing that shutting down unpermitted natural gas turbines would threaten national secur…
Socket Threat Research identified shai_hulululud@1.0.48596, an npm package designed to probe AI-based malware scanners using prompt injection, token flooding, and obfuscated JavaScript. The package contains policy-trigge…
The Senate's FY2026 National Defense Authorization Act includes bipartisan amendments banning AI from autonomously launching nuclear weapons or using lethal force without human authorization. The provisions, introduced b…
Vercel launched Enterprise Apps and Agents, a platform that enforces security and access controls for internal AI applications and agents, including identity-based authentication, short-lived credentials, and lifecycle m…