,,ii #aaith ,,congress #ast ,session ,s4 #bjha ,,in ,,the ,,senate ,,of ,,the ,,united ,,states ,june #ab1 #bjbe,ms4 ,lummis introduced the following bill2 which was read twice and referred to the ,committee on ,commerce1 ,science1 and ,transportation ,a ,,bill ,to establish immunity from civil liability for certain artificial intelligence developers1 and for other purposes4 #a4 ,short title ,this ,act may be cited as the ,responsible ,innovation and ,safe ,expertise ,act of #bjbe or the ,,rise ,act of #bjbe4 #b4 ,findings ,congress finds the following3 7#a7 ,artificial intelligence systems have rapidly advanced in capability and are increasingly being deployed across professional services1 including healthcare1 law1 finance1 and other sectors critical to the economy4 7#b7 ,industry leaders have publicly acknowledged the development of increasingly powerful artificial intelligence systems1 with some discussing the potential for artificial general intelligence and superintelligence that could fundamentally reshape the society of the ,united ,states4 7#c7 ,the current lack of clarity regarding liability for artificial intelligence errors creates uncertainty that impedes the responsible integration of these beneficial technologies into professional services and economic activity4 7#d7 ,many artificial intelligence systems operate with limited transparency regarding their capabilities1 limitations1 and default instructions1 making it difficult for professional users to assess appropriate use cases and for legal systems to fairly allocate responsibility when errors occur4 7#e7 ,learned professionals who utilize artificial intelligence tools in serving clients have professional obligations to understand the capabilities and limitations of the tools they employ1 requiring access to clear information about system specifications and performance characteristics4 7#f7 ,establishing clear standards for artificial intelligence transparency1 coupled with appropriate liability frameworks1 will promote responsible innovation while ensuring that the benefits and risks of artificial intelligence systems are properly understood and managed as these technologies continue to advance4 7#g7 ,the development of artificial intelligence systems that may significantly impact the future of human civilization warrants a governance approach that balances innovation incentives with robust transparency requirements and appropriate allocation of responsibility among developers1 professional users1 and other stakeholders4 #c4 ,definitions ,in this ,act3 7#a7 ,artificial intelligence ,the term artificial intelligence has the meaning given the term in section #ejjb of the ,national ,artificial ,intelligence ,initiative ,act of #bjbj 7#ae ,u4,s4,c4 #idja74 7#b7 ,client ,the term client means a person that-- 7,a7 engages the services of a learned professional2 7,b7 relies upon the expertise1 judgment1 and advice of the learned professional2 and 7,c7 has a relationship with the learned professional that is governed by professional standards1 codes of conduct1 or regulations4 7#c7 ,developer ,the term developer means a person that-- 7,a7 creates1 designs1 programs1 trains1 modifies1 or substantially contributes to the creation or modification of an artificial intelligence product2 7,b7 exercises control over the design specifications1 functionality1 capabilities1 limitations1 or intended uses of an artificial intelligence product2 or 7,c7 markets1 distributes1 licenses1 or makes available an artificial intelligence product under their own name1 brand1 or trademark1 regardless of whether the person creates the original underlying technology of the artificial intelligence product4 7#d7 ,error ,the term error means-- 7,a7 any output1 action1 recommendation1 or material omission by an artificial intelligence product that is false1 misleading1 fabricated1 deceptive1 or incomplete in a manner that a reasonable developer could foresee would cause harm2 or 7,b7 any failure of an artificial intelligence product to perform a function or task that the artificial intelligence product expressly or implicitly represents itself as capable of performing4 7#e7 ,learned professional ,the term learned professional means an individual who-- 7,a7 possesses specialized education1 training1 knowledge1 or skill in a profession2 7,b7 is licensed1 certified1 or otherwise authorized by an appropriate ,federal or ,state authority to practice in that profession2 7,c7 is bound by professional standards1 ethical obligations1 and a duty of care to clients2 and 7,d7 exercises independent professional judgment when using tools1 including artificial intelligence products1 in the course of rendering professional services4 7#f7 ,model card ,the term model card means a publicly available technical document in which a developer describes1 consistent with industry standards and as rigorously as or more rigorously than industry peers1 the training data sources1 evaluation methodology1 performance metrics1 intended uses1 limitations1 and risk mitigations1 including detection1 evaluation1 management1 and safeguards against errors1 of an artificial intelligence product4 7#g7 ,model specification ,the term model specification-- 7,a7 means the text or other configuration instructions of an artificial intelligence product-- 7i7 supplied by a developer2 7ii7 that establish the intended base behavior1 tone1 constraints1 or goals of the artificial intelligence product2 and 7iii7 that materially influence the outputs of the artificial intelligence product across users or sessions1 including the system prompt provided to the model before engaging with user queries2 and 7,b7 includes-- 7i7 the system prompt and any other text or images that the artificial intelligence product receives that are not visible to the end user2 7ii7 any constitution or analogous guiding document used when training or finetuning of an artificial intelligence product1 including in automated schemes in which an artificial intelligence system trains another artificial intelligence system2 and 7iii7 the instructions1 rubrics1 or other guidance provided to human raters or evaluators of an artificial intelligence product the feedback of whom is used to train or fine-tune the artificial intelligence product4 #d4 ,conditional immunity from civil liability for artificial intelligence developers 7a7 ,safe harbor eligibility ,a developer shall be immune from civil liability for errors generated by an artificial intelligence product when used by a learned professional in the course of providing professional services to a client if the developer-- 7#a7 prior to deployment of the artificial intelligence product1 publicly releases and continuously maintains-- 7,a7 the model card for the artificial intelligence product2 and 7,b7 the model specification for the artificial intelligence product1 which may include redactions-- 7i7 only relating to information that would reveal trade secrets unrelated to the safety of the artificial intelligence product2 and 7ii7 only if the developer furnishes contemporaneously with each redaction a written justification for the redaction identifying the basis for withholding the information as a trade secret2 and 7#b7 provides clear and conspicuous documentation to learned professionals describing the known limitations1 failure modes1 and appropriate domains of use for the artificial intelligence product4 7b7 ,scope of immunity ,the immunity provided under subsection 7a7 shall be conferred to a developer only for acts or omissions that do not constitute recklessness or willful misconduct by the developer4 7c7 ,duty ,to update ,immunity under subsection 7a7 relating to an artificial intelligence product shall not apply to a developer-- 7#a7 that does not update the model card1 model specification1 and documentation with respect to the artificial intelligence product as described in subsection 7a77#a7 by the date that is #cj days after the date on which the developer-- 7,a7 deploys a new version of the artificial intelligence product2 or 7,b7 discovers a new and material failure mode affecting the artificial intelligence product2 and 7#b7 of which the failure to make an update described in paragraph 7#a7 by the applicable date described in that paragraph proximately causes a harm occurring after that date4 7d7 ,preemption 7#a7 ,express preemption ,this section shall apply to any claim arising under ,state law against a developer for an error arising from the use of an artificial intelligence product by a learned professional in providing professional services if the developer is immune from civil liability under subsection 7a74 7#b7 ,claims not preempted ,nothing in this section shall apply to a claim arising under ,state law against a developer based on fraud1 knowing misrepresentation1 or conduct outside the scope of professional use of an artificial intelligence product by a learned professional4 #e4 ,preservation of other immunities and privileges ,nothing in this ,act shall be construed to affect any immunity from civil liability established by ,federal or ,state law or available at common law that is not related to the immunity established under section #d7;a74 #f4 ,effective date2 applicability ,this ,act-- 7#a7 shall take effect on ,december #a1 #bjbe2 and 7#b7 shall apply to acts or omissions occurring on or after the date described in paragraph 7#a74