Full Summary
Overnight on Monday, the US government is moving to significantly increase oversight of powerful AI models, with multiple sources including Tech Xplore, cio.com, and MSN confirming that major tech companies will now provide early access to their advanced AI for government safety testing before public release. This initiative, spearheaded by the Commerce Department's Center for AI Standards and Innovation, comes after an AI model called Mythos, developed by Anthropic, reportedly uncovered thousands of vulnerabilities in operating systems and web browsers during testing. The White House, according to Tech Xplore and MSN, objected to Anthropic giving wider access to Mythos, fearing misuse by bad actors. Companies like Google, Microsoft, and xAI are now joining OpenAI and Anthropic in this pre-release review process. In other AI news, Eurasia Review reports that the UK-led OpenBind initiative has released its first public dataset and AI model for drug discovery, accelerating the search for treatments for diseases like COVID-19 and cancer. Meanwhile, Tether, known for its stablecoin, has launched new QVAC MedPsy medical AI models that, according to The Crypto Times and Crypto Briefing, outperform Google's larger models while running directly on devices like smartphones for enhanced privacy. This push for government AI oversight means future AI releases will likely face stricter scrutiny, potentially impacting the speed of new AI innovations but aiming to make them safer for everyone. For you, the user, this could mean more secure and reliable AI in your daily life, from your smart devices to your medical care, but potentially with a slower rollout of cutting-edge features.