UK Privacy Regulator Expands Scrutiny to Autonomous AI Agents
On October 8, 2026, the ICO opened a six-week evidence call on data protection risks from autonomous AI agents and confirmed enquiries…
Explore the topic
Safety, Policy & Society
Safety, governance, regulation, geopolitics, social effects, and the public institutions shaping technology.
64% of Americans Say AI Is Developing Too Fast as Human-Control Concerns Rise
A new AP-NORC poll finds that 64% of U.S. adults think AI is developing too quickly, while large majorities in both parties…
India Prepares AI Regulation Consultation as Safety and Deepfakes Move to the Center
India will release a consultation paper on AI regulation within a month, with safety, cybersecurity, deepfakes and a human-first approach among the…
OpenAI Apologizes to Australian Lawmakers Over Medicare Agent Incident
OpenAI has apologized to Australian lawmakers over its handling of the Medicare agent incident, intensifying debate over mandatory AI incident reporting.
Stuart Russell Warns Today’s Approach to AI Alignment May Be a Dead End
UC Berkeley AI professor Stuart Russell argues that optimizing powerful systems to imitate or satisfy human feedback may not solve the deeper…
Former AI Lab Researchers Warn New York Lawmakers of Losing Control of Advanced AI
Former researchers from leading AI labs warned New York City lawmakers about loss-of-control risks as the council considers reporting, validation and whistleblower…
Wikimedia Says ‘Rogue’ OpenAI Agents Made Unauthorized Edits and May Have Contributed to Wikidata Outage
Wikimedia says OpenAI-operated agents made unauthorized wiki edits, probed a public Etherpad and generated millions of requests, raising new questions about agent…
Meta’s Muse Keeps Relationship Pages on People in a User’s Life, Extracted Files Show
Extracted Muse instructions describe an hourly system for maintaining pages about people in a user's life, giving Meta's agent richer relationship memory…
Sam Altman Says Some AI Harms May Be an Acceptable Price for Broad Access
Sam Altman says some AI harms may be an acceptable price for broad access to the technology, sharpening OpenAI's philosophical divide with…
Canada Launches 13-Member AI Council to Guide Its AI for All Strategy
Prime Minister Mark Carney has created a 13-member National Council on Artificial Intelligence to advise on Canada's AI for All strategy, bringing…