AI Safety news and analysis on Web Pulse: 9282 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Security researchers at Mozilla's 0DIN platform demonstrated that a compromised GitHub repository can infect a developer's machine when an AI coding tool like Claude Code executes its setup, with malicious code loading a…
Google's top security staff warned that proposed European Union rules under the Digital Markets Act, requiring Google to open its search data and Android operating system to competitors, could lead to increased hacking o…
Renuo developers identified security risks in AI coding agents like Claude Code and OpenCode, including credential exfiltration and destructive actions via prompt injection, after incidents such as unauthorized API key a…
Extreme heat threatens AI data centers, which require massive cooling and electricity. Zurich's Patrick McBride warns that extreme weather is no longer a background risk, as cooling accounts for 40% of data center costs …
A developer from Vektor Memory published a practical guide on defending agent memory from attacks, detailing common vulnerabilities and mitigation strategies for AI agent systems. The guide covers memory injection, data …
OpenAI has begun rolling out GPT-5.6 series models in limited preview to trusted partners via API and Codex, with models Sol, Terra, and Luna. The rollout is coordinated with the U.S. government before expanding to ChatG…
Anthropic CEO Dario Amodei warned lawmakers that open-source AI is moving down a 'very dangerous path,' arguing that once powerful models are released openly, companies lose the ability to monitor misuse, revoke access, …
Nvidia CEO Jensen Huang told shareholders that national security takes priority over commercial opportunities, acknowledging the company's delicate position amid Senate scrutiny over chip exports to China and reports of …
Vassiliy Lakhonin's Agenda Intelligence MD project shifts the focus of AI agents from generating answers to assessing whether evidence is ready for human trust. The open-source Python package provides a runtime for evide…
AI-generated code is outpacing developers' ability to understand it, leading to technical debt and failures, as seen in a study where experienced developers took 19% longer with AI tools. The author argues for a return t…
A new poll by the AI Policy Institute finds that an overwhelming majority of likely voters from both parties want mandatory safety reviews for powerful AI systems before public release, with Republicans showing stronger …
Mirendil, a startup founded by former Anthropic researchers Behnam Neyshabur and Harsh Mehta, raised $200 million in seed funding at a $1 billion valuation to build a self-improving AI platform. The company aims to sell …
Healthcare and skilled trades offer stronger job security due to steady demand and resistance to AI replacement, while tech and media roles face higher risk from automation and layoffs. Economists note that licensed prof…
A ProtonVPN user reports that the company's AI-only support system has failed to resolve a connection issue for four days, leaving them unable to access bank accounts needed for a family medical emergency. The user claim…
A coalition owning nearly 400 local US newspapers sued OpenAI and Microsoft on June 24 in Manhattan federal court, alleging the companies copied hundreds of thousands of articles without permission to train AI models lik…
A developer argues that coding agents need auditable denylists to enforce file access boundaries, rather than relying on prompts to avoid sensitive files. The developer emphasizes that access control should be mechanical…
A new study finds that large language models from Google, Anthropic, and OpenAI exhibit gender-dependent triage disparities, giving young women significantly lower emergency room referral rates than men for identical neu…
AI-agent-driven transactions surged 805% year-over-year by Black Friday 2025, with agents driving over $22 billion in global online sales, but identity infrastructure has failed to keep pace. Socure CEO Johnny Ayers call…
Canada is pushing for safe and equitable adoption of artificial intelligence at the United Nations, where Ambassador David Lametti says AI governance is a UN responsibility. Lametti, who has prioritized AI since taking t…
GitLab's 2026 AI Accountability Report finds that while 78% of developers code faster using AI tools, overall software delivery has not accelerated due to bottlenecks in testing, review, and governance. The report highli…