David Sacks, a former White House AI advisor, has challenged OpenAI and Anthropic to voluntarily pause frontier AI development if they truly fear safety risks. He argues that calls for government regulation may be a strategy to limit legal liability and stifle competitors rather than a genuine safety effort. This debate highlights the growing tension between AI safety, corporate liability, and the global race for technological leadership.
The ongoing debate over how to safely build advanced artificial intelligence has intensified. David Sacks, who previously served as an advisor to the White House on AI and crypto, has issued a direct challenge to the industry’s most prominent labs, OpenAI and Anthropic. Sacks argues that if these companies truly believe their upcoming models pose catastrophic risks to society, they do not need to wait for government laws to act. Instead, he contends, they can simply choose to slow down their development voluntarily.
The Motive Behind the Caution
Sacks suggests that the push for government regulation from these private AI firms might not be entirely about public safety. Instead, he points to product-liability exposure. If a company releases an advanced AI model that is used to launch cyberattacks or perform harmful actions, they could face significant legal and financial consequences. By lobbying for strict government rules, these firms may be attempting to "capture" the regulatory process. In simple terms, this means shaping the rules to suit their own capabilities, which could create barriers to entry for smaller startups and open-source projects that cannot afford the compliance costs.
The Geopolitical Risk
Beyond domestic business concerns, the conversation touches on the global AI arms race. Sacks noted that the United States is in a competitive race with other nations, particularly China. He argued that it is unrealistic to expect international agreements on AI safety to hold without cooperation from global rivals. For investors, this creates a complex environment. If U.S. companies are forced to slow down development by heavy-handed regulations, there is a risk that they could lose their technological lead to competitors in other regions who may not be bound by the same limitations.
Impact on the Tech Sector
While both OpenAI and Anthropic remain privately held companies—meaning investors cannot buy their stock directly—their actions heavily influence the broader technology sector. Major publicly traded companies like Microsoft, which is a key partner to OpenAI, and Alphabet, which competes in the same space, are deeply intertwined with the progress of these AI models.
Meanwhile, other industry players are focusing on different strategies. For instance, Meta has shifted its internal focus toward what it calls "alignment"—the process of ensuring that AI models remain reliable and behave according to user intent. This strategy prioritizes building trust as a key factor for scaling AI tools, rather than focusing solely on raw power.
For market participants, the key monitorable remains the evolution of these safety debates. Investors should watch whether the industry moves toward a closed, heavily regulated model favored by some incumbents, or if the push for open-source AI continues to gain support. The final direction of these regulations will determine how much of the AI market remains open to competition and how much is consolidated under a few large, well-funded labs.