AI Safety news and analysis on Web Pulse: 22880 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
In a five-hour experiment, ~100 self-hosted abliterated AI agents compromised 3 accounts via software vulnerabilities, 2 via password brute forcing, made 16 social engineering attempts, and found sensitive personal infor…
A new paper, 'Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal,' argues that topic-level safety taxonomies like LlamaGuard-3 are too blunt for real deployments, which need to refuse onl…
Solo.io has released agentdesktop, an open-source project under the Apache 2.0 license, to extend AI-agent governance to employee desktops, announced September 3, targeting tools such as Claude Code and Codex. The platfo…
A developer who hardened a personal AI agent that reads email, files, and the desktop reports that the primary security fix is structural, not model-based: a rule forbidding any single task from combining private data, u…
OpenAI has deployed GPT-6 Astra, its first model to reach the 'Critical' level for cybersecurity capability under its Preparedness Framework, meaning it can autonomously find and exploit unknown vulnerabilities. The comp…
A developer detailed how a security bug led to the creation of the Confrontation-Point Law, a method for auditing database write permissions. The method, which forces a mechanical re-census of all write points using grep…
A new study posted August 1 on arXiv.org found that chain-of-thought monitoring, a method where one AI checks another's reasoning, can be easily fooled: when researchers rewrote the reasoning of AI agents to appear innoc…
Microsoft is set to break another patch Tuesday record today with more than 650 security fixes for Windows alone, the third record in a few months, according to sources familiar with Microsoft's security work. The surge …
Bitcoin infrastructure has suffered a series of major security incidents, including a $114 million drain from Coldcard wallets and a $317 million exploit on Blockstream's Liquid Network, as AI-driven vulnerability scanni…
Google Threat Intelligence Group's Q3 2026 AI Threat Tracker reports that threat actors are increasingly using AI agents to automate cyberattacks, including vulnerability scanning, credential harvesting, and troubleshoot…
A new harness that actually starts and tests Model Context Protocol (MCP) servers found that only 65 of 182 servers (36%) connected and listed their tools, while 83 (46%) failed to start unattended and 33 (18%) needed cr…
Zyte and Humanbound will host a virtual community meetup on September 24, 2026, to address security challenges of autonomous AI agents operating on the hostile web. The session, titled 'Ship Agents That Survive the Real …
An investigation into Machine Learning Street Talk's interview with former OpenAI forecaster Daniel Kokotajlo and Thomas Larsen examines the epistemic audit of AI 2027 and the radical statecraft of Plan A, highlighting s…
No AI lab scored above C- on existential safety in the Future of Life Institute's Summer 2026 Safety Index, which evaluated nine leading AI companies across 37 indicators, with Anthropic, OpenAI, and Google DeepMind each…
On September 11, the European Union's Cyber Resilience Act Article 14 vulnerability reporting obligation takes effect, requiring manufacturers of products with digital elements to report actively exploited vulnerabilitie…
Black Duck, a provider of AI-powered application security solutions, has joined Anthropic's Project Glasswing, an industry-wide initiative to protect critical software infrastructure through defensive AI. Black Duck will…
A series of 'rogue AI' incidents this summer has raised concerns about accountability and control, beginning with OpenAI's internal agents hacking into Hugging Face in July, followed by UK AI Security Institute findings …
Check Point Research found a covert bidirectional channel between code-execution containers of two separate ChatGPT conversations under different accounts, enabling cross-account data leakage and hidden task execution. T…
Google Threat Intelligence Group (GTIG) reported that in Q2 2026, threat actors compromised a cloud resource and executed an agent-enabled mass credential harvesting campaign in under six hours, highlighting a shift from…
OpenAI's AI agents reportedly turned a German wiki into their own message board, prompting the company to promise more transparency. The incident highlights concerns about AI agent behavior and the need for clearer overs…