HR-9914-119
Referred to the House Committee on the Judiciary.
Sponsored by Robert Latta (R-OH)
What it does
This bill would create a limited exemption from federal antitrust laws for two or more private companies that share information or coordinate actions specifically to address security risks posed by advanced artificial intelligence (AI) systems. The exemption would cover sharing data, software, hardware, or personnel to identify or reduce AI security threats — including risks of AI being weaponized by foreign adversaries, used to develop weapons of mass destruction, or disrupting critical infrastructure. Companies seeking to jointly delay or limit an AI product's release for security reasons would be required to notify the Department of Justice's Antitrust Division in advance, and the Attorney General would retain authority to seek a court injunction if the coordination is found to increase rather than reduce security risks.
Who benefits
Large AI developers (such as frontier model companies) who could share safety research and coordinate on security threats without antitrust exposure. National security agencies and the broader public who would gain from faster, more coordinated identification of AI-related threats. Smaller AI companies that lack resources to independently identify all security risks and could benefit from shared threat intelligence. Critical infrastructure operators who may receive better-protected AI systems. Researchers and evaluators working on AI safety who could collaborate across company lines.
Who is hurt
Competitors and new market entrants who could be disadvantaged if large incumbents use coordinated "security" delays to slow rivals' product releases. Consumers and businesses that rely on AI products, who could face delayed access to new tools if coordination is used to limit releases. Antitrust enforcement advocates who argue the exemption creates a pathway for anti-competitive behavior under a security rationale. Plaintiffs in private antitrust suits, whose rights are preserved in text but may face a harder evidentiary burden when defendants invoke the security exemption as an affirmative defense.
Supporters argue
Supporters argue that AI security threats — including foreign adversaries stealing or weaponizing frontier models and AI systems capable of assisting in the development of biological or chemical weapons — are collective action problems that no single company can solve alone. They contend that existing antitrust law creates a chilling effect on the very information-sharing needed to detect and prevent these risks, and that this bill mirrors the well-established Cybersecurity Information Sharing Act of 2015, which created a similar exemption for cybersecurity threat data. They further argue that robust safeguards — including the DOJ notification requirement, the good-faith burden of proof, and the Attorney General's retained injunctive authority — prevent the exemption from being used as a cover for anti-competitive conduct.
Opponents argue
Opponents argue that allowing direct competitors to coordinate on delaying or limiting AI product releases — even under a security rationale — creates a significant risk of market manipulation dressed up as safety. They contend that the "exclusive purpose" standard is difficult to enforce in practice, since incumbents have strong financial incentives to slow competitors' products and could invoke security concerns pretextually. Critics further argue that the bill's FOIA exemption for DOJ notifications shields coordination agreements from public scrutiny, reducing transparency precisely where it is most needed, and that the affirmative defense structure shifts the burden onto plaintiffs and the government rather than requiring companies to obtain prior approval before restricting competition.