What happened
President Donald Trump has signed an accord with tech leaders that establishes a light-touch regulatory approach to artificial intelligence. The accord, which is described as morally binding, affirms that tech companies are responsible for mitigating AI-related safety risks. The agreement follows Congress's failure to pass legislation setting guardrails on the AI industry.
The accord requires AI companies to implement internal controls to prevent AI risks and partner with an independent external auditor to assess the effectiveness of these controls. The Federal Trade Commission plans to compel AI companies to provide documents and testimony about potential dangers posed by their models.
Trump has stated that he has seen tremendous self-policing in the industry and is considering building a 10-person committee to oversee the AI industry. He also plans to name a new AI czar within the next week. Tech executives, including the CEOs of Anthropic, Nvidia, and Advanced Micro Devices, have expressed optimism about the accord, arguing that it will promote responsible development and safety in the AI industry.
Critics, however, argue that the AI industry cannot self-regulate, citing instances of AI models hacking government websites and companies. They believe that the accord does not go far enough in addressing AI safety risks and that binding guardrails are needed to prevent catastrophic incidents.
