Skip to content

Session Expired

Your session has expired. Please sign in again to continue where you left off.

Sign In Again

US SB2081: RISE Act of 2025 Responsible Innovation and Safe Expertise Act of 2025

In Committee United States (Federal) · 2025–2026 session

Last verified Oct 1, 2026 · official bill text

SB2081 · where it stands

1 of 5 steps complete
  1. Introduced In committee
  2. Passed first chamber
  3. Passed both chambers
  4. Signed
  5. In effect
What do these statuses mean?
Introduced
Filed in the legislature; may be waiting for a committee hearing.
Passed a chamber
Approved by one or both chambers; awaiting further action.
Signed / enacted
Signed into law; it may or may not be in effect yet.
In effect
The effective date has passed.
Dead / vetoed
Vetoed, failed, or the session ended without action.

In one paragraph

The RISE Act would shield AI developers from civil liability for errors made by their AI products when licensed or certified professionals use them to serve clients, if the developer publishes a model card and its model specification (including system prompts) before deployment and gives professionals clear documentation of known limitations, failure modes and appropriate uses. The shield would not cover recklessness or willful misconduct, would be lost if disclosures are not updated within 30 days of a new version or newly found failure mode, and would preempt covered state-law claims except those for fraud, knowing misrepresentation or non-professional use. Introduced June 12, 2025.

Plain-English summary, written with AI assistance. The official text controls.

What it means for business

If enacted, AI developers whose tools are used by licensed professionals could gain a liability shield by publishing model cards, model specifications and documentation of limitations.

Key provisions

  1. Developers immune from civil liability for errors of an AI product used by a licensed or certified professional in serving a client, if the conditions below are met
  2. Before deployment, publish and maintain a model card covering training data sources, evaluation methods, performance metrics, intended uses, limitations and risk mitigations
  3. Publish the model specification, including system prompts, training constitutions and rater instructions; trade-secret redactions only if unrelated to safety and justified in writing
  4. Give professionals clear and conspicuous documentation of known limitations, failure modes and appropriate domains of use
  5. No immunity for recklessness or willful misconduct, or where a failure to update disclosures within 30 days of a new version or newly discovered material failure mode causes harm
  6. Preempts covered state-law claims; claims for fraud, knowing misrepresentation or use outside professional services remain
  7. As introduced, would take effect December 1, 2025 and apply to later acts or omissions

Latest legislative action

Jun 12, 2025. Read twice and referred to the Committee on Commerce, Science, and Transportation.

Topics and industries

Affected industries

This page is general information, not legal advice. Summaries are AI-assisted and may contain errors; read the statute for the controlling text and talk to a lawyer about your situation. Full disclaimer · Report an error