AI Fraud Accountability Act
- Bill Number
- H.R. 7786
- Origin Chamber
- House
- Congress
- 119th Congress, Session 2
- Policy Area
- Commerce
- Status
- Introduced
- Latest Action
- 2026-03-04: Referred to the Committee on Energy and Commerce, and in addition to the Committees on the Judiciary, Science, Space, and Technology, and Foreign Affairs, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned.
- Last Updated
- 2026-04-14T08:05:49Z
AI-Generated Summary
Purpose of the Legislation
The AI Fraud Accountability Act aims to protect individuals from fraud involving digital impersonations, such as AI-generated videos, audio, or images that mimic real or fictional people to deceive others. It creates criminal and civil penalties for using these technologies to scam people out of money or valuables, while promoting detection methods and international cooperation.
Key Provisions
- Criminal Prohibition (Section 2): Amends the Communications Act of 1934 to make it illegal to use "digital impersonations" (AI or tech-created visuals/audio that look or sound real) in interstate or foreign communications to impersonate someone with the intent to defraud. Exceptions apply to authorized law enforcement or intelligence activities. Penalties include fines, up to 3 years in prison, or both; threats to commit such fraud carry the same penalties. Courts must order forfeiture of profits or tools used in the crime. The law applies extraterritorially (outside the U.S. if it affects Americans).
- Civil Prohibition and FTC Enforcement (Section 3): Mirrors the criminal ban but treats violations as unfair or deceptive business practices under the Federal Trade Commission Act. The FTC can investigate, sue, and impose civil penalties, using its full authority to enforce the rules without limiting other powers.
- Working Group on Detection and Prevention (Section 4): The National Institute of Standards and Technology (NIST), under the Commerce Secretary, must form a working group within 30 days of enactment. It includes experts from DOJ, FTC, law enforcement, private sectors (e.g., finance, tech, social media), and AI/digital forensics specialists. The group develops best practices for recognizing, detecting, preventing, and tracing fraudulent digital impersonations. NIST hosts public workshops, publishes a report within 1 year, updates it annually, and reports progress to Congress yearly. The group sunsets after 10 years.
- International Cooperation (Section 5): The FTC, with input from the Attorney General and Secretary of State, identifies the top 10 foreign countries where most digital impersonation fraud originates that harms U.S. residents. The FTC can negotiate cooperation agreements with those countries' law enforcement, subject to existing legal requirements, and reports annually to Congress on progress and challenges. The DOJ reviews and updates its international agreements every 5 years to aid enforcement of cross-border crimes, with reports including recommendations.
- Savings Clause (Section 6): Ensures the law does not limit parody, satire, journalism, or other First Amendment protections (the U.S. Constitution's guarantee of free speech).
Significant Changes to Existing Law
- Adds a new subsection (i) to Section 223 of the Communications Act of 1934, expanding it from harassment and obscene calls to include AI-driven fraud via digital impersonations; also updates defenses to cover this new provision.
- Integrates enforcement into the Federal Trade Commission Act, allowing civil actions alongside criminal ones, which broadens remedies beyond just fines and jail time to include FTC-led consumer protection measures.
- Introduces mandatory interagency collaboration and international pacts, which were not previously required for this type of tech-enabled fraud.
Potential Impacts
- On Citizens: Provides stronger safeguards against scams using deepfakes (AI-altered media), potentially reducing financial losses from impersonation fraud (e.g., fake videos of celebrities or family members tricking people into sending money). It may increase public awareness through NIST's best practices.
- On Government Agencies: Expands roles—FCC gains criminal jurisdiction, FTC handles civil enforcement, DOJ pursues prosecutions and international ties, NIST leads technical research, and State Department aids diplomacy. This could strain resources but improve coordinated responses to AI threats.
- On International Relations: Fosters cooperation with foreign governments to combat cross-border fraud, potentially strengthening U.S. ties with allies on tech crime while highlighting tensions with non-cooperative nations. Annual reports may influence foreign policy on digital security.
Main Stakeholders Affected
- Individuals and Consumers: Primary victims of fraud, gaining protections but possibly facing more scrutiny in online interactions.
- Law Enforcement and Regulators: Federal (DOJ, FTC, FCC), state, and local agencies must investigate and enforce, with new tools for detection.
- Private Sector: Tech companies (e.g., AI developers, social media platforms), financial services, healthcare, retail, and telecom firms must adopt best practices, potentially increasing compliance costs but reducing fraud liability.
- Researchers and Experts: AI and digital forensics specialists contribute to the working group, influencing industry standards.
- Foreign Governments and Agencies: Targeted for cooperation, affecting diplomatic and enforcement partnerships.
Notable Legal, Constitutional, or Political Implications
- Legal: Closes gaps in existing fraud laws by specifically targeting AI technologies, making prosecution easier for "indistinguishable" fakes while allowing civil suits for faster remedies. Forfeiture procedures follow drug law models, ensuring assets from crimes are seized.
- Constitutional: The savings clause explicitly preserves First Amendment rights, avoiding challenges by distinguishing fraudulent intent from protected expression like satire—reducing risks of overbroad censorship claims.
- Political: Addresses rising concerns over AI misuse (e.g., election interference or scams), signaling bipartisan support (introduced by Republicans and Democrats). It promotes voluntary best practices over mandates, balancing innovation with security, but the 10-year sunset for the working group allows future reevaluation amid evolving tech.
This summary was generated by AI and may contain inaccuracies. Refer to the official source document for the authoritative text.
Sponsor
Cosponsors (2)
Rep. Soto, Darren [D-FL-9], Rep. Van Drew, Jefferson [R-NJ-2]
Recent Actions
- 2026-03-04: Referred to the Committee on Energy and Commerce, and in addition to the Committees on the Judiciary, Science, Space, and Technology, and Foreign Affairs, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned.
- 2026-03-04: Referred to the Committee on Energy and Commerce, and in addition to the Committees on the Judiciary, Science, Space, and Technology, and Foreign Affairs, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned.
- 2026-03-04: Referred to the Committee on Energy and Commerce, and in addition to the Committees on the Judiciary, Science, Space, and Technology, and Foreign Affairs, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned.
- 2026-03-04: Referred to the Committee on Energy and Commerce, and in addition to the Committees on the Judiciary, Science, Space, and Technology, and Foreign Affairs, for a period to be subsequently determined by the Speaker, in each case for consideration of such provisions as fall within the jurisdiction of the committee concerned.
- 2026-03-04: Introduced in House
- 2026-03-04: Introduced in House
Bill Versions
- AI Fraud Accountability Act — issued 2026-03-04 — PDF (15 pages)