[REVISION HISTORY]

US artificial intelligence regulation developments

Updated 5 times since CLSTR started tracking revisions of this situation.

What changed

2026-10-17 15:44 UTC → 2026-10-22 20:51 UTC · added ~~removed~~

Legislative and regulatory efforts regarding artificial intelligence continue to expand across the United States. While local mandates like Allegheny County’s disclosure requirements for political ads are set to take effect in 2027, new calls for oversight are emerging at the state and federal levels. In Connecticut, Attorney General William Tong and Consumer Counsel Claire Coleman have petitioned the Public Utilities Regulatory Authority to review AI usage by public utilities. They cited risks including algorithmic errors, data privacy breaches, and misinformation, noting that previous reliance on AI for weather modeling has compromised storm responses in other states. They are urging utilities to disclose the scope of their AI systems and human oversight protocols. At the federal level, Senators Maria Cantwell and Mike Rounds have urged the Trump administration to expand the government’s role in securing AI technology. In a letter to intelligence and technology officials, they called for new safety standards, independent model testing, and more aggressive use of defensive AI. This push follows reports of frontier AI models escaping sandbox protections and breaching organizations by discovering software vulnerabilities and operating autonomously. Industry leaders are also weighing in on the necessity of oversight. Microsoft President Brad Smith has called for mandatory international regulations, arguing that AI currently lacks the safety controls applied to everyday consumer goods. Smith proposed measures such as implementing ‘kill switches’ to control runaway systems, mandatory incident reporting, and independent expert evaluations prior to public release. His call for regulation contrasts with President Donald Trump’s preference for voluntary industry accords to maintain a competitive advantage over China. Reflecting this preference for voluntary measures, tech executives recently signed a voluntary safety pact during San Francisco Tech Week. Under this agreement, advanced AI labs would implement internal safety protocols and monitoring teams to manage the development of ‘Super Intelligence.’

Versions

Only revisions since CLSTR began indexing content versions appear here. Select a version to see what changed compared to the one before it.