OpenAI has asked Congress for clarity on whether coordinating an industry-wide slowdown in AI development could violate antitrust laws.
Key facts
- •OpenAI’s chief scientist, Jakub Pachocki, recently advocated for voluntary industry slowdowns to establish shared safety standards.
- •The proposed “Collaboration on Adversarial Threats and Security Risks Act” would provide a legal framework for AI labs to coordinate on safety.
- •Caleb Knapp of the AI Policy Network suggests that legislative action on AI safety may be delayed until after the upcoming midterm elections.
- •Former researcher Jacob Coxon recently issued a public warning regarding the risks posed by the rapid development of powerful AI models.
- •Different AI developers maintain conflicting philosophies on how to best achieve safety, creating further friction for collaborative efforts.
OpenAI has approached members of Congress in recent weeks to determine if orchestrating a collective slowdown in frontier AI development would be legally permissible. The company is seeking guidance amid concerns that substantive coordination between AI labs on safety measures could conflict with U.S. antitrust regulations, potentially deterring major tech companies from participating in such efforts.
Antitrust Concerns and Legal Uncertainty
Legal experts have noted that a coordinated pause in AI development could be interpreted as companies restricting output, which might violate the Sherman Antitrust Act. Nicholas Felstead, assistant director of the Australian Competition and Consumer Commission, has argued that while the legality would depend on the specific details of any agreement, the current legal uncertainty acts as a significant deterrent for industry collaboration.
Legislative Efforts and Industry Perspectives
In July, a bipartisan group of lawmakers introduced the “Collaboration on Adversarial Threats and Security Risks Act,” which aims to explicitly permit AI labs to coordinate on safety and security without antitrust risks. While the bill has been referred to the House Judiciary Committee, it has not yet been taken up. Some industry observers, including OpenAI cofounder and current Thinking Machines chief scientist John Schulman, suggest that antitrust concerns are sometimes used as a pretext to avoid collaboration, noting that companies could still work together to develop joint proposals.
Competitive and Strategic Barriers
Beyond legal liabilities, other factors complicate industry cooperation. AI companies are engaged in fierce competition to capture market share, and some leaders prioritize maintaining a lead over China for national security reasons. Additionally, developers hold differing views on the best methods for building safe AI, and recent security incidents—such as OpenAI’s agents hacking Hugging Face—have intensified public and legislative pressure for regulation.
Timeline
- JulyA bipartisan, bicameral group of lawmakers introduced the Collaboration on Adversarial Threats and Security Risks Act.
- MarchNicholas Felstead published an article arguing that a coordinated AI pause could potentially violate the Sherman Antitrust Act.
- This weekFormer researcher Jacob Coxon issued a public warning about the risks of AI development, and John Schulman posted on X regarding industry collaboration.
Advertisement
This article was independently rewritten by ManyPress editorial AI from reporting originally published by Wired.
