S 3680, the *Eliminating Bias in Algorithmic Systems Act of 2026*, requires federal agencies using or funding complex algorithms (like AI systems) to establish civil rights offices focused on bias. These offices must monitor and address algorithmic bias related to protected characteristics (e.g., race, gender, disability) that affect government programs or economic opportunities. Agencies must submit biennial reports to Congress detailing bias risks, mitigation steps, stakeholder engagement, and recommendations, and join an interagency working group on algorithmic fairness. The bill directly affects covered federal agencies, not private companies or the general public.
This bill creates a new grant program to increase U.S. artificial intelligence research capacity by funding specific institutions. It provides competitive funding to institutions not in the top 100 for federal research spending, historically Black colleges, minority-serving institutions, Tribal Colleges, or consortia of these entities. Funds can be used for AI research programs, faculty development, student bridge programs, research resources, and integrating ethical AI practices. The program aims to broaden participation in AI from underserved communities and avoid duplicating existing federal AI initiatives.
The FUTURES Act (S 3855) establishes a formal U.S.-Israel Defense Technology Cooperation Initiative to accelerate joint development and integration of defense technologies. It directs the U.S. Secretary of Defense to identify Israeli-origin technologies for rapid adoption into American military systems, focusing on areas like counter-drone systems, missile defense, AI, cyber security, and directed energy. The bill authorizes $150 million annually (2027-2029) for this initiative, requiring regular reports to Congress on progress, technology transitions, and industry partnerships. This policy directly affects U.S. defense contractors, Israeli defense firms, and military acquisition programs by creating new pathways to incorporate Israeli innovations into U.S. systems.
The Artificial Intelligence Scam Prevention Act makes it illegal to use AI to impersonate people for fraudulent purposes, such as creating fake voices or images to scam consumers. It requires clear disclosure when AI is used in phone calls or text messages, so people know they're interacting with technology rather than a real person. The bill establishes an advisory group with FTC, FCC, and industry representatives to develop best practices for preventing these scams and mandates regular reports on AI-enabled fraud. The law aims to protect consumers from increasingly sophisticated scams, which cost over $1.4 billion in 2024 alone.
The Artificial Intelligence Civil Rights Act of 2025 regulates AI systems that make decisions affecting people's lives, such as employment, housing, healthcare, credit, and education. It requires developers and deployers to conduct pre-deployment evaluations and annual impact assessments by independent auditors to prevent discrimination and harm. The law mandates clear consumer disclosures about AI use, provides the right to human alternatives for AI-driven decisions, and creates enforcement mechanisms through the Federal Trade Commission and state attorneys general. It also includes whistleblower protections for employees reporting violations and requires transparency about how AI systems operate.
The Algorithmic Accountability Act of 2025 requires companies that deploy automated decision systems for critical areas like employment, housing, healthcare, and financial services (known as "covered entities") to conduct impact assessments before and after deployment. These assessments must evaluate fairness, privacy, security, and potential negative impacts on consumers, particularly for protected groups like racial minorities, women, and people with disabilities. Covered entities must document these assessments and submit annual summary reports to the Federal Trade Commission, which will make anonymized versions publicly available in a searchable online repository. The law applies to companies with over $50 million in annual revenue or those handling data for over 1 million consumers. The goal is to increase transparency and accountability around automated decision systems that significantly affect people's lives.
The MedShield Act of 2025 establishes a new pandemic preparedness program under the Department of Health and Human Services (HHS), requiring the use of artificial intelligence to improve national biodefense. It directs HHS to create a system for real-time pathogen tracking, accelerate vaccine development, and optimize treatment distribution using AI tools, as outlined in Section 3(c). The program mandates annual funding of $300 million in 2025, increasing to $500 million by 2029 (Section 6), and requires a report to Congress within 180 days of enactment (Section 4). This bill directly affects federal agencies coordinating pandemic response and aims to create a continuous, integrated system for addressing biological threats, replacing ad-hoc efforts like Operation Warp Speed.
HR 6306, the AI Fraud Deterrence Act, increases penalties for fraud crimes when artificial intelligence is used to impersonate federal officials. The bill amends existing laws (mail fraud, wire fraud, bank fraud, and money laundering) to add $1 million fines or up to 20 years in prison for AI-assisted fraud, and specifically targets impersonation of government officials with $1 million fines or up to 3 years in prison when AI is used. It defines "artificial intelligence" using the 2020 National AI Initiative Act and includes a First Amendment protection for satire or parody with clear disclosure. The law directly affects scammers using AI to mimic government officials, as highlighted by recent incidents involving impersonated White House and State Department communications.
The QUIET Act (HR 1027) requires that robocalls or text messages using artificial intelligence to mimic human communication must disclose at the beginning that AI is being used. It doubles penalties for violations involving AI impersonation intended to defraud, cause harm, or obtain value, applying to automated calls sent to stored numbers or generated randomly. The bill excludes calls requiring substantial human intervention, focusing on transparency and deterrence in AI-driven automated communications.
This bill aims to increase U.S. involvement in global technical standards for artificial intelligence and emerging technologies. It requires the National Institute of Standards and Technology (NIST) to provide Congress with a briefing on U.S. participation opportunities, create a public web portal listing international standards efforts, and launch a 5-year pilot program. The pilot program will offer grants (up to $5 million total) to U.S. organizations hosting international standards meetings, covering costs like venue and planning. These efforts target U.S. industry and federal agencies working on AI/tech standards, with annual reports to Congress and a requirement to propose permanent implementation if the pilot succeeds.
