TRAIN Act
- Bill Number
- H.R. 7209
- Origin Chamber
- House
- Congress
- 119th Congress, Session 2
- Policy Area
- Commerce
- Status
- Introduced
- Latest Action
- 2026-01-22: Referred to the House Committee on the Judiciary.
- Last Updated
- 2026-07-08T20:06:15Z
AI-Generated Summary
Purpose
The legislation establishes a new administrative subpoena process under U.S. copyright law. This process allows copyright owners to obtain information from developers about whether their protected works were used to train generative artificial intelligence models.
Key Provisions
- Short Title: The bill is named the Transparency and Responsibility for Artificial Intelligence Networks Act, or TRAIN Act.
- New Section Added: It inserts Section 514 into Chapter 5 of Title 17 of the United States Code, creating rules for subpoenas related to AI model training.
- Definitions: The bill defines key terms, including:
- Artificial intelligence (referencing existing federal law).
- Artificial intelligence model (a system using computational or machine-learning techniques to produce outputs).
- Developer (a person or government agency that designs, produces, or modifies a generative AI model and handles its training data; excludes noncommercial end users).
- Generative artificial intelligence model (a model that creates synthetic content like images or text from input data).
- Substantially modify (actions that materially change a model's function, such as retraining).
- Training material (individual works or components used to train a model, including text, images, or annotations).
- Subpoena Request Process:
- A copyright owner (or authorized representative) with a good-faith belief that their works were used in training may file a request with a U.S. district court clerk.
- The request includes a proposed subpoena and a sworn statement confirming the belief, the purpose (to check for use of their works), and that records will only support protecting their rights.
- The subpoena is limited to the requester's own works and cannot seek information about others' copyrights.
- Issuance and Compliance:
- If the request is properly formatted, the clerk issues the subpoena, ordering the developer to disclose copies or identifying records of the training materials.
- Developers must respond promptly.
- Recipients of the information must keep it confidential unless authorized to share.
- Enforcement and Remedies:
- Procedures follow the Federal Rules of Civil Procedure for subpoenas, with remedies for noncompliance.
- Failure to comply creates a rebuttable presumption that the developer copied the copyrighted work.
- Courts may impose sanctions for bad-faith requests, applying Rule 11 of the Federal Rules of Civil Procedure.
- Effective Date: The section takes effect upon enactment.
Significant Changes to Existing Law
This bill introduces a targeted subpoena mechanism for disclosing AI training data, which does not currently exist in copyright law. It creates a streamlined administrative process outside of full litigation, while incorporating elements of civil procedure rules. No other amendments to Title 17 are made beyond adding this section and updating the table of contents.
Potential Impacts
- On Government Agencies: Federal district courts would handle subpoena requests, potentially increasing administrative workload for clerks and judges.
- On Citizens: Copyright owners gain a tool to investigate potential unauthorized use of their works, which could aid in protecting intellectual property. AI developers face new disclosure obligations that may affect business operations and data practices.
- On International Relations: The bill applies to U.S. developers and courts but does not address foreign entities or cross-border data issues directly.
Main Stakeholders Affected
- Copyright owners and their representatives.
- Developers of generative AI models (including companies and agencies involved in training).
- U.S. district courts responsible for issuing subpoenas.
- Users and beneficiaries of AI models who may indirectly be affected by changes in training practices.
Notable Legal, Constitutional, or Political Implications
The bill operates within the existing framework of copyright enforcement and civil procedure, raising questions about balancing intellectual property rights with developers' interests in proprietary training data. It includes safeguards like good-faith requirements and sanctions to prevent misuse, but the rebuttable presumption could influence future disputes over AI-related copying. The process is designed to be neutral and court-supervised, without altering broader constitutional standards for subpoenas or copyright protections.
This summary was generated by AI and may contain inaccuracies. Refer to the official source document for the authoritative text.
Sponsor
Cosponsors (2)
Rep. Moran, Nathaniel [R-TX-1], Rep. Lee, Laurel M. [R-FL-15]
Recent Actions
- 2026-01-22: Referred to the House Committee on the Judiciary.
- 2026-01-22: Introduced in House
- 2026-01-22: Introduced in House
Bill Versions
- Transparency and Responsibility for Artificial Intelligence Networks Act — issued 2026-01-22 — PDF (8 pages)