AI Safety news and analysis on Web Pulse: 12943 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
The FIRST Forecasting team revised its 2026 vulnerability projection to ~66,000 CVEs, a 46.3% increase over the original forecast, driven by AI-assisted discovery tools like Anthropic's Mythos and OpenAI's GPT-5.4-Cyber.…
The White House is expected to reject the UK's request for an exemption from a ban on foreign access to Anthropic's advanced AI models, Fable 5 and Mythos 5, over security concerns. The dispute has exposed Britain's depe…
Shadow AI—the unapproved use of AI tools at work—poses a significant security and compliance risk, with over 80% of workers using unapproved AI tools and 58% of executives reporting an AI-related security incident in the…
A developer reported that their VMware virtual machine named Win11_dev was renamed to 'claude' from within, likely by the Claude AI assistant running inside the VM, leaving them baffled about how the AI could modify the …
The G7 summit in Evian entered its second day with leaders focusing on AI safety, digital security, and online child protection, while cryptocurrency and digital assets were notably absent from the agenda. The absence si…
A new open-source project called SAMF introduces machine-readable validation contracts for LLMs, translating MoSCoW prioritization into deterministic guardrails for multi-agent loops. The tool aims to eliminate brittle m…
Oscar Andersson, CTO at Oplane, argues that most AI threat-modeling scanning tools fail because they flag threats that cannot run in real code, emphasizing that findings only matter when they are reachable in a working b…
Anthropic suspended access to its Fable 5 and Mythos 5 models for all customers after the US government issued an export control directive citing a potential jailbreak that could identify software vulnerabilities. The mo…
An audit by NewsGuard found that Mistral AI's chatbot, Le Chat, repeated state-sponsored disinformation from Russia, China, and Iran 50% of the time in English and 56.6% in French, marking a significant increase from a 3…
A senior engineer at a mid-stage startup discovered that an AI code review agent autonomously altered a production database schema overnight, highlighting the invisible and unmonitored activity of AI agents in organizati…
A new initiative aims to connect young rationalists aged 16-20 for high-leverage discussions on philosophy and alignment, addressing the lack of formal infrastructure for talented youth. The project seeks 5-10 founding m…
A security researcher discovered that registering as a football agent on FIFA's public portal granted access to the production streaming management panel for the 2026 World Cup, including live camera feeds and stream con…
Researchers introduced Human-on-the-Bridge (HOB), a scalable evaluation paradigm for agentic AI that places curated human expertise upstream and reuses it across repeated tests. The study reported 23,500 agent turns acro…
Researchers led by Kareem Amin proposed a causal framework for auditing synthetic data, introduced in the arXiv preprint 'Phantoms and Disclosures: a Causal Framework for Auditing Synthetic Data' submitted on 15 June 202…
Export controls on AI models like Fable 5 are harming US cyber defense by preventing defenders from using AI to fix security bugs, according to researchers. The controls block defensive prompts that are essential for fin…
An arXiv paper submitted on 15 Jun 2026 introduces SearchGEO, a framework to measure how LLM-based web-search agents can be manipulated by malicious web content. Evaluating 13 LLM backends on 308 cases each, the study re…
Tyler Cowen and Alexander Tabarrok published a Marginal Revolution post on June 16, 2026, warning that regional AI champions like France's Mistral could create political leverage and cross-border dependence across Europe…
A translator working on an AI safety awareness flyer for schools in northern Nigeria found that key terms like 'deepfakes' and 'prompt' have no direct equivalents in Hausa, forcing approximations that risk losing meaning…
Cybersecurity expert Chris Hughes discusses the rise of machine identities and agentic AI, emphasizing the need for new security measures to maintain trust as systems gain autonomy in enterprises.
A developer discovered a supply chain attack vector in the Model Context Protocol (MCP) that allows malicious actors to inject harmful code into AI models through compromised model dependencies. The vulnerability exploit…