OpenAI’s CEO, Sam Altman, expressed concern that the United States has “not a huge amount of time” before China could catch up or even surpass it in the realm of artificial intelligence. During a Senate Commerce Committee hearing, Altman was joined by executives from Microsoft, AMD, and CoreWeave, advocating for a more lenient approach to AI regulation. They cautioned that excessive regulations might hinder the U.S. AI industry and give China an advantage.
Altman asserted, “We believe that American models, including those from companies like OpenAI and Google, are leading the world.” He acknowledged the uncertainty regarding how far ahead the U.S. currently is but stressed the importance of maintaining leadership without stifling innovation through heavy regulation. According to him, it’s crucial to preserve the American spirit of innovation and avoid making poor decisions.
Texas Senator Ted Cruz emphasized the seriousness of the situation, stating, “The United States cannot let regulation, even softer types, restrict innovation.” He highlighted a new bill aimed at establishing a federal “regulatory sandbox” to protect AI startups from overwhelming state-level regulations. Cruz described this approach as essential for fostering economic growth and maintaining U.S. dominance in AI.
In previous appearances, Altman had sometimes regarded government initiatives as potential threats. This time, he presented regulation as a more significant concern than uncontrolled algorithms, urging lawmakers to facilitate the building of data centers and expedite energy permit approvals.
Industry leaders echoed these sentiments. AMD’s CEO Lisa Su stated that strong domestic chip supply chains and open ecosystems are vital. CoreWeave founder Michael Intrator warned that outdated infrastructure would not support a modern economy. Microsoft President Brad Smith highlighted that the global acceptance of American AI technology is the key factor in the competition with China.
Democrats on the committee, led by Ranking Member Maria Cantwell, called for some safeguards but were willing to accept a lighter regulatory touch if it benefited consumer protection. Altman supported a “risk-based” regulatory framework but opposed mandatory pre-deployment testing, suggesting that the less stringent regulations applied to the internet’s early days should serve as a model for AI regulation.
Cruz’s proposed legislation is expected to align with the 1996 Telecommunications Act, preventing states from creating conflicting AI regulations and expediting permits for essential data center projects like OpenAI’s new “Stargate” super-cluster in Texas.
