AI Safety news and analysis on Web Pulse: 9616 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
OpenAI, Anthropic, Amazon and Microsoft are funding RAISE US, a new nonprofit led by former US commerce secretary Gina Raimondo to retrain American workers for the AI economy. The group has raised over $500 million with …
A new bipartisan nonprofit, RAISE US, launches with over $500 million to retrain American workers for the AI era, partnering with states and major employers like Amazon and Microsoft to prevent catastrophic job losses.
ARENA (Alignment Research Engineer Accelerator) announced its ninth iteration, a 4-5 week ML bootcamp focused on AI safety, running in-person at LISA in London from October 5 to November 6, 2026. Applications are due by …
A security researcher developed Lattice Mind, a hybrid AI security agent combining deterministic scanners with LLM intelligence, which solved the hardest challenges from PicoCTF 2025 and DawgCTF 2026 in under 30 minutes …
A new anxiety called 'hybrid FOMO' is emerging as people fear being left behind in the partnership between natural and artificial intelligence, according to a Psychology Today article. The phenomenon combines fear of AI …
Scaled Cognition raised $100 million in Series A funding led by Khosla Ventures to build AI models that do not hallucinate. The startup's APT model is designed with reliability engineered into its architecture, targeting…
Anthropic accused Alibaba of orchestrating the largest known 'distillation attack' on its Claude AI model, using 25,000 fraudulent accounts to conduct 29 million exchanges between April 22 and June 5. The attack aimed to…
Postman argues that production-grade AI agents depend more on data quality, API reliability, and execution governance than on model intelligence. The company reframes agent building as a systems engineering challenge, em…
Eon co-founder Gonen Stein warns that AI agents can delete production data in seconds, while most recovery systems still operate at human speed. Eon's research found 98% of executives confident in recovery despite most s…
The European Union has joined the US-led Pax Silica initiative, a multinational pact to secure AI and semiconductor supply chains, expanding transatlantic tech cooperation. The non-binding agreement covers critical miner…
Anthropic CEO Dario Amodei was replaced by co-founder Tom Brown in White House meetings after officials found Amodei difficult to engage with, according to a report. The change comes as the Trump administration imposed e…
An Azure AI agent passed all functional, retrieval, and safety tests but still requires human approval before executing actions. The developer refused full autonomy because enterprise accountability must be designed befo…
A researcher published a follow-up post compiling novel AI safety research and project ideas from the past 10 months, including proposals for live databases tracking research questions and interventions, and analyses of …
The European Union held talks with the White House to regain access to Anthropic's advanced AI models after a US export-control order cut off European access in mid-June. The order, citing a jailbreak concern, forced Ant…
Nearly 1,000 actors, agents, and parents signed an open letter organized by the Agents for Young Performers Association condemning a major studio's contract clauses that require child actors to allow their voices to be u…
Insurers are using generative AI diffusion models to generate thousands of synthetic weather events for catastrophe modeling, improving risk assessment for rare disasters. However, the models suffer from hallucinations t…
AI security firm AISLE discovered six new CVEs in curl, including the oldest latent bug in the project's history, and previously found all 12 zero-day vulnerabilities in OpenSSL's latest security release. The findings de…
Europe's largest tech conference VivaTech in Paris drew 180,000 attendees amid a heatwave, with Jeff Bezos predicting AI will create labor shortages and Yann LeCun warning of a potential AI bubble. The event focused on c…
The x401 protocol extends HTTP with a cryptographic identity assertion layer, enabling servers to require verified identity before granting access and agents to prove human authorization. As AI agents surpass human web t…
Anthropic disabled global access to its Fable 5 and Mythos 5 AI models after a US export control directive, prompting EU leaders to push for a 'trusted partners' framework to restore allied access. The move has disrupted…