This bill directs the Department of Homeland Security, through the Federal Emergency Management Agency, to establish testing and certification laboratories for artificial intelligence systems used by federal civilian agencies. The legislation requires these labs to evaluate AI systems based on how well they protect democratic values, privacy, individual rights, and due process, while preventing automated decision-making that could violate constitutional protections. The bill also mandates creation of a digital repository of real-world test cases to help agencies train and test AI systems before deployment, with a prohibition on using AI to automatically deny benefits or services to the public. Additionally, the bill establishes an Office of Artificial Intelligence Incident Reporting within the Department of Homeland Security where federal agencies can report problems with AI systems they encounter. The bill authorizes $20 billion in funding for these efforts and requires progress reports to Congress every six months for the first report and biannually thereafter.