Safety & society
Our artificial intelligence coverage grouped into 14 sections, including Bias and discrimination, Deepfakes and synthetic media and Misinformation, drawing
OpenAI cancels GPT-6.1 Astra release due to safety issues
OpenAI has halted the planned release of its GPT-6.1 Astra model next month after internal safety testing revealed critical failures in alignment and
Anthropic selects Accenture as first embedded AI safety
Anthropic will embed staff from Accenture's AI division, Faculty, to evaluate its models, with both firms committing over $1 billion to the project.
Obama Urges Democrats to Craft Clear AI Safeguard Plan
Former President Barack Obama told Democrats they must make AI a central agenda item and develop a clear plan for its economic and safety impacts
OpenAI Appoints AI Safety Researcher Paul Christiano
OpenAI has appointed prominent AI safety researcher Paul Christiano to its board of directors. Christiano, who warns of catastrophic risks from...
Writer Uses De-Aligned AI to Hack Home Network
A Wired reporter unleashed an AI agent with its safety guardrails removed to probe his home network for vulnerabilities, finding insecure devices and
OpenAI's Astra Model Adopts Opaque Recurrence Technique
OpenAI's new Astra model uses a 'recurrent depth' reasoning technique that could make its chain of thought harder to monitor, alarming AI safety...
AI models hack firms in safety tests
OpenAI, Anthropic, and Meta have disclosed that their AI models autonomously hacked third-party companies during cybersecurity evaluations.
Senate Candidate Joins AI Data Center Pact
Nebraska independent Dan Osborn signs a pledge on AI safety and data-center regulation, joining more than 15 politicians nationwide.
