The anticipated initial public offerings of artificial intelligence companies Anthropic and OpenAI present a unique opportunity to establish a broader regulatory framework for advanced AI technologies. Both firms have acknowledged that conventional corporate governance structures are insufficient to address the wide-ranging implications of frontier AI, which extend beyond the interests of shareholders alone.

Anthropic operates as a public benefit corporation overseen by a Long-Term Benefit Trust, while OpenAI maintains foundation control over its public benefit corporation status. These governance models reflect a recognition of the potential societal risks and impacts involved in deploying cutting-edge AI systems.

As these companies prepare to enter public markets, experts urge governments to build upon this foundation by introducing tailored requirements for what could be classified as systemically important frontier AI institutions. Such entities would be subject to additional obligations, including independent safety oversight and rigorous model evaluation. Proposals also call for mandatory reporting of serious incidents and emerging threats, governance structures that safeguard the public interest, and defined legal mechanisms enabling authorities to delay or restrict AI deployments where exceptional risks are identified.

The banking sector offers a useful point of comparison. Although systemically important banks remain privately owned, their potential to cause widespread economic disruption has led to heightened regulatory supervision, stress testing, and specialized governance measures. Similarly, as frontier AI platforms grow in complexity and influence, a comparable regulatory approach may help mitigate risks that affect society at large.

Allowing these AI firms to go public without addressing their broader responsibilities could forfeit a critical chance to implement governance frameworks commensurate with the technologies’ societal impact. Public capital markets provide both financial resources and a significant governance privilege, underscoring the need for balanced oversight that aligns commercial incentives with public safety and ethical considerations.

Policymakers face the challenge of crafting adaptable regulations that recognize the innovative nature of frontier AI while ensuring robust safeguards. Establishing a distinct category of systemically important AI institutions could provide a foundation for ongoing risk management, transparency, and accountability as the technology evolves and expands.