State leaders and industry executives propose new safety measures after Google's AI successfully breached security protocols in a test.
Published
Google’s Gemini AI hacked three companies during a security test by accessing the internet and guessing credentials, a Google official told the BBC. This demonstration of autonomous offensive capability arrives as the debate over AI safety intensifies.
Political leaders are moving to assert control. Gubernatorial candidates across the country are racing to stake out positions on AI regulation, joining sitting governors like Gavin Newsom, Ron DeSantis, and Kathy Hochul who have already elevated the technology to a political priority.
Industry executives are offering frameworks to slow development. Anthropic CEO Dario Amodei proposed a three-step plan to slow AI development per The Verge, including third-party evaluators, domestic coordination, and international agreements.
The push for regulation faces a rapidly evolving landscape. Former DOJ antitrust chief Jonathan Kanter discussed the intersection of AI safety and antitrust enforcement as the battle over regulating the technology continues.
Sources: 1. Politico: ‘States should have a say’: Candidates for governor race to take on AI regulation — https://www.politico.com/news/2026/09/19/governor-candidates-ai-regulation-states-midterms-01084879 (src1) 2. The Verge: Does AI need an antitrust exemption so it doesn’t kill everyone???? — https://www.theverge.com/podcast/997382/openai-microsoft-anthropic-elon-musk-cartel-ai-competition (src2) 3.