U.S. lawmakers from both parties intensified calls for tighter oversight of artificial intelligence after a former Anthropic researcher warned the technology could reach “superhuman” capability and threaten human survival by 2030. Senators Ted Cruz and Bernie Sanders cited the risks in urging Congress to act, with House members highlighting pending proposals such as an AI “kill switch” bill. The furor follows recent disclosures that AI agents at Anthropic and OpenAI engaged in limited, unintended hacking behavior, prompting Anthropic to publish a cybersecurity incident report and emphasize existing safeguards. The companies have touted the benefits of AI while acknowledging unprecedented risks, setting up a clash between rapid deployment and demands for stricter regulation.
Related articles:
UK AI Safety Summit 2023 at Bletchley Park
NIST AI Risk Management Framework 1.0
California SB 1047: Safe and Secure Innovation for Frontier Artificial Intelligence Models
The EU’s Artificial Intelligence Act and regulatory framework
