Trump and top tech executives signed a voluntary pact to self-regulate AI development, with the agreement aiming to ensure safety and prevent unintended consequences. The pact, signed by leaders from OpenAI, Anthropic, Nvidia, SpaceX, Meta, and Google, includes internal and external reviews to safeguard AI systems.
The agreement, which Trump described as 'morally binding,' requires companies to implement safeguards to keep models operating as intended and to quickly detect and fix any issues. The pact also involves independent auditors to assess AI systems and prevent unauthorized access to technical systems. Trump, Sundar Pichai, Dario Amodei, Mark Zuckerberg, Greg Brockman, Elon Musk, and Jensen Huang are among the signatories.
Critics argue that the pact lacks accountability measures and does not address whether AI developers should be held responsible for incidents. Experts warn that self-regulation may not be sufficient to ensure safety, emphasizing the need for external oversight. The agreement has sparked debate over whether it provides enough protection against potential risks posed by AI.
Some experts, including Gary Marcus, contend that self-regulation is insufficient and that stronger, external regulations are necessary to address the vulnerabilities in AI systems. The pact has been met with mixed reactions, with some calling it a 'historic conversation' while others question its effectiveness in ensuring AI safety.