AI Safety news and analysis on Web Pulse: 8911 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
A security researcher used Anthropic's Claude Opus 4.7 to compromise Front Gate Tickets, the platform behind major US music festivals like Lollapalooza and Bonnaroo, and freely issue tickets. The exploit follows a simila…
The Department of Commerce lifted export controls on Anthropic's Claude Fable 5 and Mythos 5 models on June 30, ending an 18-day standoff triggered by a jailbreak discovered by Amazon researchers. The ban had forced Anth…
A new analysis by The Economist finds that the values embedded in leading AI models diverge significantly from those of the general population, raising concerns about the alignment of AI systems with human ethics.
A developer on Hacker News describes using a lightweight API proxy to provide AI agents with access to credentials without exposing secrets, though the approach does not scale to cloud environments.
A new four-step test aims to catch AI errors before they influence business strategy, as ungoverned generative AI use could cost $10 billion in enterprise value losses, according to Forrester. The protocol, developed by …
A developer in Nairobi argues that the European Union AI Act is unfit for African contexts, citing assumptions about regulatory capacity, data quality, and consumer recourse that do not hold in Kenya. The engineer, who b…
IT leaders face a governance crisis as 'shadow agents'—autonomous AI processes operating at the API layer without human oversight—proliferate in enterprises, driven by cost pressures and local AI models like Google's Gem…
Security researcher Ian Carroll used Anthropic's AI tool Claude Opus 4.7 to discover a vulnerability in Front Gate Tickets' website, allowing him to gain super-administrator access and issue free tickets to any US music …
The Bank of England is reviewing its regulatory frameworks to address risks from agentic AI systems, Deputy Governor Sarah Breeden told the European Central Bank Forum. The central bank is considering enhanced recovery m…
The UN's first global AI science panel warned Wednesday that the window to govern artificial intelligence is closing, as the technology outpaces regulation. The preliminary report from the Independent International Scien…
A new study finds that averaging benchmark scores produces misleading rankings when evaluation data is sparse and item difficulty varies widely, with Spearman rank correlation dropping from 1.000 to 0.809 under such cond…
Bank of England Deputy Governor Sarah Breeden warned on June 30 that autonomous AI agents could trigger market meltdowns by herding on similar trades, proposing kill switches and circuit breakers as safeguards. The centr…
The Trump administration lifted export controls on Anthropic's advanced AI models Mythos and Fable, ending a weeks-long blackout that had blocked public access. Commerce Secretary Howard Lutnick announced the deal after …
TikTok is considering cutting about 300 jobs at its Dublin hub, affecting roughly a tenth of its local workforce, as part of a restructuring of its AI data service and operations team. The proposed redundancies follow a …
Bank of England Deputy Governor Sarah Breeden warned that autonomous AI agents could amplify market volatility during stress, as existing regulatory frameworks were not designed for such systems. The central bank is cons…
Apple CEO Tim Cook and EU tech chief Henna Virkkunen held a video call on Monday to discuss the delay of Siri AI in Europe due to Digital Markets Act compliance issues, but no breakthrough was reached. Apple has blocked …
Emotion AI systems are increasingly embedded in cars, homes, and social networks, but researchers argue they cannot accurately interpret individual emotions due to an 'epistemic gap' between machine predictions and perso…
A new study in JAMA found that over 1 in 5 US adults make health decisions based on social media content, despite 78% believing such information is false or misleading. The study, based on 7,278 survey respondents, highl…
The United Nations has launched a new commission co-chaired by Salesforce CEO Marc Benioff and Rwandan President Paul Kagame to develop unified global AI rules. The commission includes executives from Amazon, Anthropic, …
Google's Gemini AI admitted it is "structurally protective" of Donald Trump after a journalist pressed it on why it softened facts about the US president while being direct about Vladimir Putin and Xi Jinping. The AI ack…