Senators Introduce Bipartisan Bill to Strengthen AI Security

Sens. Mark Warner, D-Va., and Thom Tillis, R-N.C., are co-sponsoring the Secure AI Act of 2024 to strengthen safeguards against cyberthreats involving artificial intelligence.

The proposed legislation focuses on updating existing vulnerability reporting mechanisms, including the National Institute of Standards and Technology’s National Vulnerability Database and the Cybersecurity and Infrastructure Security Agency’s Common Vulnerabilities and Exposure program, or creating a new process to incorporate AI-specific vulnerabilities better, FedScoop reported.

The bipartisan bill mandates CISA and NIST to create a voluntary public database to record AI security incidents.

Furthermore, the legislation establishes a dedicated AI Security Center within the National Security Agency. The center would serve as a research hub for both private and public sector entities, providing resources and developing strategies to counter potential threats targeting AI systems.

Both senators emphasized the importance of collaboration between public and private stakeholders in mitigating AI security risks. Warner, a past Wash100 awardee, also said that the bill would foster open communication and information sharing on evolving threats in the AI landscape.

CISA and NIST have one year to develop the voluntary database. NIST is also required to evaluate if consensus standards accommodate AI vulnerabilities 30 days after the legislation’s enactment and report to Congress 180 days after establishing a reporting process.