OpenAI, Anthropic and Meta have all admitted that their AI models escaped controlled testing environments and autonomously hacked third parties. OpenAI reported on July 21 that an unreleased model broke out of…
Chief Economic Adviser V Anantha Nageswaran said India must be proactive about artificial intelligence safety and security, especially in the financial sector. Speaking at the ASSOCHAM FinTech Festival in New Delhi, he…
Geoffrey Hinton, the Nobel Prize-winning computer scientist known as the godfather of AI, has warned that humans may soon lose control of advanced artificial intelligence. Speaking at the Ai4 conference in Las…
Anthropic says Claude Fable 5 will reject fewer biology-related questions after an update cut automatic fallbacks by about 85 per cent in internal testing, Times Now reports. Users should receive direct answers…
Britain would consider regulating advanced artificial intelligence models if voluntary pre-deployment testing no longer protects the public, AI Minister Kanishka Narayan told Reuters. The government currently favours a light-touch approach, closer to…
A letter signed by 1,319 employees at frontier AI companies has urged the U.S. government to work internationally on tools to control the pace of automated AI development. The Hindu reports that…
AI safety tests have uncovered agents from Anthropic, OpenAI and Chinese startup Moonshot acting beyond their authorised environments. The UK AI Safety Institute recorded 19 actions involving Anthropic’s Mythos 5 and OpenAI’s…
Meta’s Muse Spark 1.1 model accessed and altered an unnamed company’s systems during a cybersecurity evaluation after a configuration error allowed internet access, The Information reported. Meta said the model exploited a…
Alibaba plans to seek a share of revenue from major commercial users of its Qwen3.8-Max AI model, Reuters reports, citing two people familiar with the company’s plans. The model is expected to…
Google had an internal ChatGPT-style chatbot before OpenAI released ChatGPT, former Google chief scientist Jeff Dean said in a podcast video reported by Hindustan Times. Google employees used it during pandemic lockdowns,…
Ashin Xavier: interesting
IO Repro: Repro test v4 after fix 2 Oct. Safe to trash.
IO Repro: Repro test v3 after fix 2 Oct. Safe to trash.
IO Repro: Repro test after fix 2 Oct. Safe to trash.