AI Safety news and analysis on Web Pulse: 10947 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Cloudflare, major browser vendors including Mozilla Firefox, Google Chrome, and Microsoft Edge, and Shopify are developing the Private Access Control Tokens (PACT) protocol to replace invasive bot detection with anonymou…
Semantic caching in AI systems reuses conclusions from similar queries, reducing latency and cost, but introduces risks in agentic systems where cached answers can trigger tool calls, permissions, or deletions without th…
Anthropic updated its privacy policy to require age or identity verification for Claude users under certain circumstances, sparking concerns about digital surveillance and government overreach. Critics worry the move cou…
Brennan, an Indigenous blogger, writes an open letter to compassionate, left-leaning meat eaters who oppose generative AI, urging them to consider veganism as a way to reduce harm. He acknowledges the environmental and s…
Scammers are using deepfake AI to impersonate people in video calls, making it harder to detect fraud. The technology enables realistic fake identities, posing a growing threat to personal and financial security.
AI companies including OpenAI and Anthropic are pouring millions into US midterm elections through super PACs, spending $37 million so far on both Republican and Democratic candidates. The funding aims to crush non-parti…
Oracle cut 21,000 jobs (13% of its workforce) in the past year, citing the adoption of AI technologies as a direct cause. The software giant is spending $70 billion on AI infrastructure in 2026, joining other tech compan…
A major outage at Anthropic on June 23, 2026, disrupted Claude AI services, highlighting the risks of relying on a single LLM provider. The incident, caused by demand outpacing infrastructure capacity, followed similar d…
Google Cloud announced new Confidential Computing innovations, including Confidential G4 VMs with NVIDIA RTX PRO 6000 Blackwell GPUs, to enable verifiable private AI inference and fine-tuning globally. The service uses h…
The Alphabet Institute has secured a physical facility at Cavendish University in Lusaka, Zambia, to host weekly Effective Altruism and 80,000 Hours workshops starting July 4. Founder, who also works with Zambia's Presid…
GitHub joined an open source coalition including Black Forest Labs, Hugging Face, and Mozilla Corporation to advocate for amendments to California's AI Transparency Act (SB 942/SB 1000), arguing that its license revocati…
Mount Sinai Health System in New York has adopted Signal 1's AI management platform to monitor and optimize its AI tools, marking a major partnership for the Toronto-based startup. The move comes as healthcare providers …
A security expert warns that AI agents are creating a new software supply chain risk by fetching and running unvetted open-source packages in ephemeral sandboxes, bypassing traditional scanners and review processes. The …
Meta suffered its most embarrassing outage ever when a bug allowed users to change any account's email by simply asking the Meta AI bot, leading to high-profile account takeovers. Engineers attributed the disaster to AI-…
Byrne Hobart questions the reliability of Hacker News commenters in his latest edition, also covering threshold effects, the open-face AI sandwich, legal DDOS, arms races, and sales funnels.
Developer NotASithLord released peerd, an open-source AI agent harness that runs entirely inside the browser as a Chrome/Firefox extension, enabling agents to read and drive web pages, spin up sandboxed compute, and comm…
Tesla admitted its Full Self-Driving system was engaged during a fatal crash in Katy, Texas, that killed a 76-year-old woman, but blamed the driver for overriding the system by pressing the accelerator to 100%. The compa…
Anthropic engineering head Fiona Fung said Claude Code made employees' work a 'lonely experience,' prompting the company to introduce hackathons and pair programming to foster team interaction. The issue reflects broader…
Anthropic's Opus 4.7 and 4.8 model upgrades improved benchmarks but broke production systems, sparking developer backlash on Reddit and X. The issue stems from benchmarks not reflecting real-world prompts, with models be…
A Desiderata roundup examines common misunderstandings about LLM consciousness, noting that philosopher Dan Dennett explicitly rejected the idea that chatbots are conscious, contrary to claims by blogger Matt Yglesias. T…