Collaboration on Adversarial Threats and Security Risks Act

Risk Mitigation Through Cooperation: The bill allows two or more non-Federal entities to exchange information or assistance regarding "covered artificial intelligence security risks" (Section 3(a)(1)). This could help protect against threats such as the theft or weaponization of AI by covered nations, the development of weapons, or disruptions to critical infrastructure (Section 2(6)).

Controlled AI Deployment: Entities can coordinate to reduce risks by delaying or limiting the release, deployment, or development of artificial intelligence (Section 3(a)(2)). This requires the entity to provide written notice to the Assistant Attorney General detailing the specific risk (Section 3(a)(2)).

Protection of Sensitive Information: Information submitted to the Assistant Attorney General regarding proposed restrictions is deemed voluntarily shared and is withheld from the public (Section 3(e)).

Burden of Proof on Entities: If an entity uses the antitrust exemption, they bear the burden of proving by a "preponderance of the evidence" that their actions were taken in good faith and for the exclusive security purpose intended (Section 3(c)).

Risk of Increased Security Threats: The Attorney General can seek an injunction against these coordinated actions if it is demonstrated that the actions are "reasonably likely to result in an overall increase in covered artificial intelligence security risks" (Section 4(b)(2)).

Strict Limitations on Information Use: A non-Federal entity receiving information or assistance must implement "reasonable internal controls" to ensure that the information is not used for any purpose other than the specific security purpose intended (Section 3(b)).