AI safety and regulation news covering governance frameworks, AI legislation, safety research, alignment, and policy debates on artificial intelligence.
Top AI leaders from OpenAI, Anthropic, Google DeepMind, and Microsoft signed a letter urging Congress to require synthetic DNA/RNA sellers to screen customers and orders to prevent misuse for biological weapons. The letter warns that AI progress could erode barriers to obtaining bioweapons.
Anthropic, Google DeepMind, and Meta have hired psychologists, philosophers, and ethicists to study machine consciousness and AI welfare. Anthropic is testing models for behaviors resembling panic and anxiety, while DeepMind hired philosopher Henry Shevlin to work on machine consciousness and AGI readiness.
Sam Altman met with White House officials and lawmakers on Capitol Hill to discuss AI regulation. Sen. Bernie Sanders announced legislation for the federal government to take a 50% ownership stake in frontier AI labs.
OpenAI released a policy paper calling for mandatory evaluations of advanced AI models by civilian agencies, opposing President Trump's voluntary NSA-led framework. CEO Sam Altman is in Washington on Wednesday to advocate for the proposal, which would give the Commerce Department's CAISI a greater role.
Australia is among 150 organisations in 15 countries granted access to Anthropic's Claude Mythos Preview, an AI model deemed too dangerous for public release. Independent MP Kate Chaney is urging the government to strengthen its AI response amid readiness concerns.
This CNN analysis argues that Waymo robotaxis, while statistically safer than human drivers, introduce new safety problems such as running red lights and failing to navigate obstacles. It contends that these incidents, documented in government records and videos, are early warning signs that regulators and the company must address before expansion.
President Trump signed an executive order scaling back review requirements for AI model developers submitting models to the federal government. The order reduces the voluntary review period from 30 days to a shorter timeframe, reportedly after lobbying by David Sacks to speed up market access.
This commentary contrasts Trump's and Sanders' AI visions, arguing that Trump's executive order appropriately avoids mandatory government licensing while Sanders' proposal would give the federal government excessive control over private AI companies. The author contends that government involvement risks political misalignment and stifling innovation.
This commentary summarizes key discussions from The Hill's Invest in America Summit, highlighting how AI is rapidly transforming the economy and how US tariffs are impacting businesses. It presents arguments from various speakers on the need for AI safeguards and the goals of trade policy, without endorsing any single perspective.
Florida Attorney General James Uthmeier filed a civil lawsuit against OpenAI and CEO Sam Altman, alleging deceptive trade practices and negligence. The suit cites ChatGPT's role in planning a mass shooting at Florida State University and the killing of two graduate students at USF, and also claims the AI promotes self-harm, cognitive decline, and behavioral addiction. Florida is the first state to file such a lawsuit.