AI Slowdown Faces Antitrust Concerns Amid Safety Debates

AI leaders are exploring a development slowdown, but antitrust laws and internal disagreements pose challenges.

OpenAI has recently sought guidance from Congress regarding the legality of an industry-wide slowdown in frontier AI development. Coordination on safety among AI labs could potentially violate antitrust law, presenting a significant hurdle for major tech companies to participate. Jakub Pachocki, OpenAI's chief scientist, suggested that coordinating to slow down future development is crucial for ensuring the safety of self-improving AI systems, anticipating voluntary slowdowns until shared safety standards are established.

However, legal scholars, including Nicholas Felstead, have indicated that a coordinated pause might be seen as restricting output, potentially conflicting with the Sherman Antitrust Act. While some lawmakers are addressing these concerns with proposed legislation like the "Collaboration on Adversarial Threats and Security Risks Act," which aims to permit AI labs to coordinate on security and safety without antitrust violations, the bill has yet to be taken up by the Judiciary Committee.

Beyond legal issues, competition for market share and differing opinions on safe AI development among industry leaders contribute to the complexity. John Schulman, an OpenAI cofounder and chief scientist at Thinking Machines, urged OpenAI and Anthropic to collaborate on a pacing proposal, dismissing antitrust as a "fake" barrier to joint development. The debate intensifies following warnings from former Anthropic and OpenAI researcher Jacob Coxon about the risks posed by AI developers.

Find the right AI tool for your business

Chat with Insta and get matched to the right tool in seconds.