The Cato Institute warns that a government-mandated pause in AI development could stifle competition and protect dominant firms. Policy scholar Jennifer Huddleston argues that such regulations might fail to improve safety while hindering future technological advancements. She suggests that companies should instead rely on voluntary safeguards and shared standards to address specific risks.
Huddleston notes that regulatory frameworks designed for current models may become obsolete, potentially preventing the development of safer alternatives. She also warns that pausing domestic AI progress could weaken cybersecurity, especially as foreign competitors continue their research. Furthermore, she argues that rules shaped by industry leaders could create barriers for smaller companies.
Jack Dorsey, chairman and co-founder of Block, echoed these concerns on X. He opposes industry-wide limits negotiated by current leaders, arguing that these measures could exclude researchers and developers who might build alternatives or expose failures. Dorsey emphasizes that protecting a company's commercial advantage should not be confused with legitimate safety objectives.
Instead of broad pauses, Dorsey advocates for open releases that allow independent researchers to inspect, modify, and test models. He supports independent testing and public enforcement, suggesting that model releases should only be withheld if they present clear, catastrophic risks that cannot be mitigated through other means. He rejects general approval requirements or waiting periods.
The debate intensifies as lawmakers, including Senator Bernie Sanders and Representative Greg Casar, propose pausing advanced AI development until federal standards are set. Meanwhile, OpenAI CEO Sam Altman has expressed support for slower development and federal rules. The Atlantic Council highlights that competitive pressures and international tensions complicate the push for enforceable safety standards.
