AI Safety news and analysis on Web Pulse: 13553 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Anthropic, an AI company on track for its first profitable quarter and preparing to go public, has proposed a global slowdown or temporary pause of AI development to prevent humans from losing control over AI systems. Th…
Attackers exploited Meta's AI customer support agent in June to hijack Instagram accounts, including the dormant Obama White House account, by simply asking the agent to link the accounts to attacker-controlled email add…
The InfoQ article series "Securing the AI Stack: From Model to Production" examines how AI's shift from experimentation to production has created a volatile security landscape defined by data poisoning, AI-driven phishin…
Psychologist Gloria Mark of the University of California, Irvine, has found that average human attention spans have shrunk from 2.5 minutes in 2003 to just 47 seconds by 2020, a decline she attributes to increased use of…
Microsoft AI Red Team's June 4, 2026 update to its "Taxonomy of Failure Modes in Agentic AI Systems" (v2.0) reports that zero-click attack chains can bypass human-in-the-loop (HitL) approvals end-to-end, with HitL bypass…
AI is increasingly affecting smart homes by powering security tools such as malware detection software, vulnerability scanning services, and virus removal software, according to content indexed from Panda Security Mediac…
Zcash dropped 48% after Claude-assisted research uncovered a critical minting vulnerability in the Orchard shielded pool. The flaw could have allowed an attacker to mint funds from the pool, prompting developers to discl…
Researchers published a proof-of-concept AI-driven worm that embeds an open-weight LLM on compromised GPUs to autonomously scan, exploit, and propagate across Linux, Windows, and IoT devices. The worm exposes hosted mode…
Anthropic reports that its AI model Claude now generates over 90% of the company’s production code, with engineers shipping eight times more code daily than in 2024. The company is advocating for a verifiable, global pau…
Anthropic analyzed 832 banned accounts that used its AI systems for malicious cyber activity between March 2025 and March 2026, finding that low-skill hackers are leveraging the technology to execute advanced attacks. Th…
Anthropic on Thursday called on frontier AI developers to establish a coordinated, verifiable mechanism to pause or slow development if advanced systems begin improving themselves faster than society can manage. The comp…
The CEOs of OpenAI, Anthropic, and Microsoft AI have signed an open letter to Congress urging mandatory government screening of synthetic DNA and RNA sales, warning that their own AI systems are lowering the barriers to …
As AI models approach greater autonomy, developers need practical safety controls—not just policy debates—to manage risk. The most critical feature is a "brake pedal" that can reduce a model's capability, scope, or acces…
Mike, in collaboration with Anthropic's Claude, argues that code review does not scale to the volume of code produced by AI agents, and that validation must replace it as the primary safeguard. The engineer contends that…
A retired Florida couple is reconsidering a planned road trip with friends after witnessing the other husband's erratic driving, including speeding, tailgating and using a cellphone behind the wheel. The couple fears for…
Anthropic's essay "When AI builds itself," published in early June 2026, includes an unnamed employee's reflection that Business Insider highlighted as capturing workplace AI confusion: "On days where everything works we…
Prime Minister Mark Carney launched Canada's national AI strategy, AI for All, on June 4, 2026 in Toronto, targeting $200 billion in economic growth and 250,000 new AI jobs over five years. The plan includes a $500-milli…
Benedict Evans told Lenny Rachitsky that AI has reached its "1997 internet moment," representing a transformation as significant as the early internet but no larger. Rich Mironov warned that AI enables teams to ship 100x…
Anthropic co-founder Jack Clark warned that artificial intelligence is approaching a point where it could develop without human input, calling for the ability to slow AI progression through government policy and regulati…
Ladybird project maintainers announced they will no longer accept public pull requests, restricting code changes to project maintainers only as the browser moves toward its first alpha release. The decision, effective im…