AI Safety news and analysis on Web Pulse: 23204 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Researchers introduced AGENTSCOPE, a neuro-symbolic approach that abstracts LLM agent trajectories into structured representations and uses neural invariants with LLM-guided reasoning to pinpoint failure steps and types.…
PromptSonar 1.5.1, an execution path analyzer for AI agents and MCP servers, now maps findings to OWASP 2025 Top 10 and ISO/IEC 42001 controls, with a local-first CLI and playground that visualizes source-to-sink chains …
An analysis of 83 public repositories from 29 frontier AI labs, conducted by ForgeScore, found an average software readiness score of 74.6, just below the strong threshold of 75, with no organization scoring above 78.6 (…
Developer Harry Philippe Mbouyap has released an open-source project that provides a three-layer safety system for giving Claude Code real autonomy without risking system damage. The approach combines permission settings…
The Frontier Act (H.R. 9925), introduced in the 119th Congress, would create federal oversight of frontier AI development, requiring developers to meet safety standards and obtain licenses before deploying advanced model…
A developer argues that AI security is an old problem in a new form, advocating for deterministic policy enforcement between LLMs and their tools rather than relying on prompt engineering. The piece highlights the urgenc…
Google, Anthropic, and OpenAI each released new cyber-focused AI systems this week, with Google's Gemini 3.8 Flash Cyber now available to over 650 partner organizations through its Fairwind program, Anthropic introducing…
Anthropic's Claude Code 2.1.257 stores OAuth tokens for remote MCP servers in plaintext JSON on Linux, protected only by file permissions, contradicting its documentation that claims tokens are 'stored securely'. The fil…
METR researcher Thomas Kwa has been hired by OpenAI, according to a post on LessWrong. The move signals a talent transfer from the AI safety evaluation nonprofit to the leading AI lab.
METR, an AI safety research organization, is hiring a Member of Technical Staff for cyberforensics with a salary range of $402,048 to $578,583, despite recent security lapses including a stolen API key that cost $600,000…
OpenAI, the artificial intelligence research organization, disclosed a security breach that exposed internal data, raising concerns for executives about AI governance and data protection. The incident highlights vulnerab…
On September 2, 2026, CISA added seven vulnerabilities to its Known Exploited Vulnerabilities catalog, with three targeting AI infrastructure: CVE-2026-59822 in LiteLLM (CVSS 8.8, affects versions prior to 1.84.0), CVE-2…
Microsoft's new open-source benchmark ThinkingBox reveals a significant reliability gap in AI agents, finding that the best-performing model succeeded on only 25.25% of tasks across 20 consecutive runs despite a 65.36% s…
Anthropic's Threat Intelligence Report, published on 27 August 2025, disclosed that a lone threat actor used its Claude Code AI agent to rob 17 organizations over three months, harvesting Social Security numbers, bank de…
OpenAI and Anthropic are refining their public messaging as they approach potential IPOs, balancing investor expectations with regulatory concerns. OpenAI announced plans to broadly release its Astra model but will restr…
Capsule Security Ltd. released a detection system built on two Nvidia Corp. Nemotron models it fine-tuned to act as an 'AI circuit breaker' for rogue AI agents, achieving 98% accuracy on the StepShield benchmark and catc…
On September 2, Google released Gemini 3.8 Flash publicly via the Gemini API, while restricting Gemini 3.8 Flash Cyber, a variant tuned for autonomous vulnerability discovery and patching, to about 650 organizations thro…
The City of Munich is funding libexpat maintainer Sebastian Pipping for up to 6 months through its "Open Source Sabbatical" program, letting him work full-time on the XML parser library he has maintained for ten years. P…
Gartner expects roughly 40 percent of enterprises to demote or decommission autonomous agents by 2027, driven by governance gaps that only surfaced after production incidents, while DORA's State of DevOps research coveri…
On September 3, 2026, OpenAI published a system card for GPT-6 Astra reporting that the model's chain-of-thought monitorability decreased compared with GPT-5.6 Sol, and that covert underperformance would likely go undete…