What's happening
This bill requires health insurers to disclose, regulate, and monitor their use of artificial intelligence in reviewing and denying medical care.
Why it matters
Directly impacts all New Yorkers with private health insurance, as well as healthcare providers statewide. , by requiring insurers to disclose when and how AI algorithms are used in making decisions about coverage for medical care—potentially increasing transparency and accountability in a process that affects millions.
Imposes new regulatory oversight and consumer protections. Insurers must regularly submit their AI algorithms and training data to state regulators for review, show efforts to minimize bias, adhere to evidence-based guidelines, and subject their systems to ongoing quality assurance testing—aiming to prevent discriminatory or unsafe denials of care.
Expands documentation and reporting requirements for denial of care. Any denial (adverse determination) influenced by AI must include detailed explanations, the credentials and time spent by the human reviewer, and a plain-language description of the AI’s role—empowering patients and providers to challenge decisions and understand the rationale behind them.
Albany, before your first meeting.
Every bill that moved yesterday, summarized in a sentence and ranked by impact. The Albany Scorecard follows on Sundays.
Free. Unsubscribe in one click. Want alerts on specific bills? Create an account.
Who cares
Patients/Insureds/Enrollees
Individuals covered by health insurance plans who may be subject to utilization review decisions involving artificial intelligence. They are directly affected by how these algorithms impact coverage, denials, privacy, and transparency regarding their care.
Health Care Providers
Doctors, hospitals, and clinicians who interact with utilization review processes and must respond to adverse determinations (coverage denials) potentially influenced by AI. This legislation impacts their administrative burden, appeals processes, and access to information about algorithmic decision-making.
Health Insurers and Health Maintenance Organizations
Companies that offer health insurance and use (or plan to use) AI-based algorithms in utilization review. They will need to comply with new transparency, audit, reporting, and data use requirements, and face penalties for violations.
Clinical Peer Reviewers
Medical professionals who make final determinations on utilization reviews, especially when AI is involved. They are held to new documentation, attestation, and compliance standards, and may be subject to fines or professional discipline for violations.
Headline styles
Disclose AI in Health Reviews
Plain English, a few words. Used as the page title.
Sponsors
No co-sponsors yet.
History
- Referred to insurance
Companion bill
The same bill in the other chamber. A bill becomes law when one version passes both houses and the governor signs it, so either number can carry it.
Disclose AI in Health Reviews
In Assembly Committee · Insurance · Sponsor David Weprin
Full text
7 pagesUnderlined is new[Struck] is removedOfficial text on nysenate.gov ↗
STATE OF NEW YORK
10241
IN SENATE
May 8, 2026
Introduced by Sen. HINCHEY -- read twice and ordered printed, and when printed to be committed to the Committee on Insurance
AN ACT to amend the insurance law and the public health law, in relation to the use of artificial intelligence for utilization review
The People of the State of New York, represented in Senate and Assembly, do enact as follows:
Section 1. The insurance law is amended by adding a new section 338 to read as follows:
§ 338. Notification regarding the use of artificial intelligence for utilization review. (a) As used in this section, the following terms shall have the following meanings:
(1) "Adverse determination" shall have the same meaning as such term is defined by section forty-nine hundred of this chapter.
(2) "Artificial intelligence-based algorithm" means any algorithmbased system that can, for a given set of human-defined objectives, make predictions, recommendations and/or decisions influencing real or virtual environments. An artificial intelligence-based algorithm uses machine and human-based inputs to:
(A) perceive real and virtual environments;
(B) abstract such perceptions into models through analysis in an automated manner; and
(C) use model inference to formulate options for information or action.
(3) "Clinical peer reviewer" shall have the same meaning as such term is defined by section forty-nine hundred of this chapter.
(4) "Emergency condition" shall have the same meaning as such term is defined by section forty-nine hundred of this chapter.
(5) "Health care provider" shall have the same meaning as such term is defined by section thirty-two hundred twenty-four-a of this chapter.
(6) "Insurer" means an insurer authorized to write accident and health insurance in this state, a corporation organized pursuant to article forty-three of this chapter, and a health maintenance organization certified pursuant to article forty-four of the public health law.
EXPLANATION--Matter in italics (underscored) is new; matter in brackets
[ ] is old law to be omitted.
LBD15313-01-6
(7) "Utilization review" shall have the same meaning as such term is defined by section forty-nine hundred of this chapter.
(b)(1) The superintendent shall require all insurers to provide written notice to insureds, enrollees, and health care providers about the use of artificial intelligence-based algorithms in their utilization review process. Such written notice shall be sent when the artificial intelligence-based algorithm is first adopted by the insurer and once during each subsequent policy period. In addition, such insurers shall provide a clear and conspicuous notice on their websites about their use of artificial intelligence-based algorithms in the utilization review process.
(2) If artificial intelligence-based algorithms are used in the utilization review process, each insurer shall disclose to the insured's or enrollee's health care provider, and, upon request, to the insured or enrollee, the following information: the criteria governing such artificial intelligence-based algorithm, the data sets used to train such artificial intelligence-based algorithm, the artificial intelligencebased algorithm itself, and the outcomes produced by software that uses the artificial intelligence-based algorithm.
(3) An insurer's use of artificial intelligence-based algorithms in the utilization review process shall be disclosed to the insured or enrollee and to the insured's or enrollee's health care provider in the notice of an adverse determination made by clinical peer reviewers pursuant to subsection (e) of section forty-nine hundred three of this chapter and subdivision five of section forty-nine hundred three of the public health law.
(4) Any artificial intelligence-based algorithms used in the utilization review process by an insurer shall not base an adverse determination solely on a group data set.
(5) Any artificial intelligence-based algorithms used in the utilization review process by an insurer shall not directly or indirectly cause harm to the insured or enrollee.
(6) Any artificial intelligence-based algorithms used in the utilization review process by an insurer shall be open to inspection for audit or compliance reviews by the superintendent in accordance with applicable state and federal laws.
(7) Insured and enrollee data shall not be used beyond its intended and stated purpose, consistent with the federal Health Insurance Portability and Accountability Act of 1996, as applicable.
(c) Every insurer shall submit the artificial intelligence-based algorithms and training data sets that are being used or will be used in the utilization review process to the superintendent. Such submissions shall include, at a minimum, a certification that such artificial intelligence-based algorithms and training data sets:
(1) have minimized the risk of bias based on the covered person's race, color, religious creed, ancestry, age, sex, gender, national origin, handicap or disability, and are consistent with state and federal antidiscrimination laws;
(2) adhere to evidence-based clinical guidelines;
(3) do not rely on information that is not in compliance with the requirements for utilization review pursuant to section forty-nine hundred three of this chapter and section forty-nine hundred three of the public health law; and
(4) do not independently create or change clinical standards or other coverage criteria.
(d)(1) An insurer with a utilization review process that initially uses artificial intelligence-based algorithms, shall ensure that adverse determinations are made only by clinical peer reviewers. A clinical peer reviewer who participates in a utilization review process for an insurer that initially uses artificial intelligence-based algorithms for a utilization review shall, prior to issuing an adverse determination: consider the requesting health provider's recommendation, the insured's or enrollee's medical or other clinical history, as applicable, and individual clinical circumstances; open and document the utilization review of the individual clinical records and data; and comply with all requirements of article forty-nine of this chapter and article fortynine of the public health law regarding utilization review. In addition, a utilization review determination with respect to an emergency condition shall require application of a prudent layperson standard to the patient's presenting symptoms, without regard to final diagnosis.
(2) Every insurer shall submit to the superintendent, in such form and manner as the superintendent may require, data on the amount of time a clinical peer reviewer spends examining an adverse determination prior to signing off on each adverse determination.
(3) An artificial intelligence-based algorithm shall not be the sole basis of an insurer's decision to deny, delay, or modify health care services based, in whole or in part, on medical necessity.
(4) Every clinical peer reviewer that participates in a utilization review process for an insurer issuing adverse determinations shall issue a signed statement in accordance with paragraph one of subsection (e) of section forty-nine hundred three of this chapter or paragraph (a) of subdivision five of section forty-nine hundred three of the public health law.
(e) Every insurer shall establish an ongoing quality assurance testing process that meets any requirements that the superintendent may specify with defined parameters on safety and efficacy of all artificial intelligence-based algorithms, and shall submit the results of such quality assurance testing process to the superintendent at such time and in such form and manner as the superintendent may specify, but not less frequently than semiannually. The results of such quality assurance testing process shall be published on a public website within thirty days of the submission of such results to the superintendent.
(f) (1) A violation of the provisions of this section shall be subject to one or more of the following penalties at the discretion of the superintendent, in consultation with the commissioner of health and the commissioner of education as applicable:
(A) Where a violation is made by an insurer:
(i) suspension or revocation of license;
(ii) refusal, for a period not to exceed one year, to issue a new license;
(iii) a fine of not more than five thousand dollars for each violation of this section; or
(iv) a fine of not more than ten thousand dollars for each willful violation of this section.
Fines imposed pursuant to the provisions of this subparagraph on a single insurer shall not exceed five hundred thousand dollars in aggregate during a calendar year.
(B) Where a violation is made by a clinical peer reviewer:
(i) suspension or revocation of license;
(ii) refusal, for a period not to exceed one year, to issue a new license;
(iii) a fine of not more than five thousand dollars for each violation of this section; or
(iv) a fine of not more than ten thousand dollars for each willful violation of this section.
Fines imposed pursuant to the provisions of this subparagraph on a single clinical peer reviewer shall not exceed one hundred thousand dollars in aggregate during a calendar year.
(2) Penalties pursuant to the provisions of this subsection shall be in addition to any other remedies or penalties that may be imposed under any other applicable law.
(g) The superintendent shall promulgate all rules and regulations necessary for the implementation of this section.
§ 2. Subparagraph (C) of paragraph 9 of subsection (i) of section 3216 of the insurance law, as amended by chapter 219 of the laws of 2011, is amended to read as follows:
(C) For purposes of this paragraph, an "emergency condition" means a medical or behavioral condition that, regardless of the final diagnosis, manifests itself by acute symptoms of sufficient severity, including severe pain, such that a prudent layperson, possessing an average knowledge of medicine and health, could reasonably expect the absence of immediate medical attention to result in (i) placing the health of the person afflicted with such condition in serious jeopardy, or in the case of a behavioral condition placing the health of such person or others in serious jeopardy; (ii) serious impairment to such person's bodily functions; (iii) serious dysfunction of any bodily organ or part of such person; (iv) serious disfigurement of such person; or (v) a condition described in clause (i), (ii) or (iii) of section 1867(e)(1)(A) of the Social Security Act.
§ 3. Item (ii) of subparagraph (E) of paragraph 24 of subsection (i) of section 3216 of the insurance law, as amended by chapter 219 of the laws of 2011, is amended to read as follows:
(ii) "Emergency condition" means a medical or behavioral condition that, regardless of final diagnosis, manifests itself by acute symptoms of sufficient severity, including severe pain, such that a prudent layperson, possessing an average knowledge of medicine and health, could reasonably expect the absence of immediate medical attention to result in (I) placing the health of the person afflicted with such condition in serious jeopardy, or in the case of a behavioral condition placing the health of such person or others in serious jeopardy; (II) serious impairment to such person's bodily functions; (III) serious dysfunction of any bodily organ or part of such person; (IV) serious disfigurement of such person; or (V) a condition described in clause (i), (ii), or (iii) of section 1867(e)(1)(A) of the Social Security Act.
§ 4. Subparagraph (C) of paragraph 4 of subsection (k) of section 3221 of the insurance law, as amended by chapter 219 of the laws of 2011, is amended to read as follows:
(C) In this paragraph, an "emergency condition" means a medical or behavioral condition that, regardless of the final diagnosis, manifests itself by acute symptoms of sufficient severity, including severe pain, such that a prudent layperson, possessing an average knowledge of medicine and health, could reasonably expect the absence of immediate medical attention to result in (i) placing the health of the person afflicted with such condition in serious jeopardy, or in the case of a behavioral condition placing the health of such person or others in serious jeopardy; (ii) serious impairment to such person's bodily functions; (iii) serious dysfunction of any bodily organ or part of such person; (iv) serious disfigurement of such person; or (v) a condition described in clause (i), (ii) or (iii) of section 1867(e)(1)(A) of the Social Security Act.
§ 5. Subparagraph (C) of paragraph 2 of subsection (a) of section 4303 of the insurance law, as amended by chapter 219 of the laws of 2011, is amended to read as follows:
(C) For the purpose of this provision, "emergency condition" means a medical or behavioral condition that, regardless of the final diagnosis, manifests itself by acute symptoms of sufficient severity, including severe pain, such that a prudent layperson, possessing an average knowledge of medicine and health, could reasonably expect the absence of immediate medical attention to result in (i) placing the health of the person afflicted with such condition in serious jeopardy, or in the case of a behavioral condition placing the health of such person or others in serious jeopardy; (ii) serious impairment to such person's bodily functions; (iii) serious dysfunction of any bodily organ or part of such person; (iv) serious disfigurement of such person; or (v) a condition described in clause (i), (ii) or (iii) of section 1867(e)(1)(A) of the Social Security Act.
Reflowed for reading: line numbers and page breaks removed. Cite the official text.
Sponsor's memo
Title of bill
An act to amend the insurance law and the public health law, in relation to the use of artificial intelligence for utilization review
Purpose
The purpose of this bill is to ensure transparency, accountability and appropriate clinical oversight in the use of AI-based algorithms by health insurers in making coverage determinations. The legislation would protect patients and providers by requiring disclosure of AI use, preventing AI from serving as the sole basis for adverse determinations, and reinforcing the role of qualified clinical peer reviewers in cover- age decisions.
Summary of provisions
Section 1 adds a new section 338 to the insurance law, establishing notification and transparency requirements, mandates additional human review of adverse determinations, requires ongoing quality assurance for the safety and efficacy, and authorizes penalties for noncompliance by health insurers that utilize artificial intelligence algorithms in their utilization review process.
Read the full memo (792 words)
Section 2 amends subparagraph (C) of paragraph 9 of subsection (i) of section 3216 of the insurance law, as amended by section 3 of chapter 219 of the taws of 2011 to clarify that the definition of "emergency condition" applies regardless of the final diagnosis.
Section 3 amends clause (ii) of subparagraph (E) of paragraph (24) of subsection (i) of section 3216 of the Insurance Law, as amended by section 7 of chapter 219 of the laws of 2011, to clarify that the defi- nition of an "emergency condition" applies regardless of the final diag- nosis.
Section 4 amends subparagraph (C) of paragraph (4) of subsection (k) of section 3221 of the insurance law, as amended by section 13 of chapter 219 of the laws of 2011 to clarify that the definition of "emergency condition" applies regardless of the final diagnosis.
Section 5 amends subparagraph (C) of paragraph (2) of subsection (a) of section 4303 of the insurance law, as amended by section 25 of chapter 219 of the laws of 2011 to clarify that the definition of "emergency condition" applies regardless of the final diagnosis.
Section 6 amends subsection (c) of section 4900 of the insurance law, as amended by section 51 of chapter 219 of the laws of 2011 to clarify that the definition of "emergency condition" applies regardless of the final diagnosis.
Section 7 amends subdivision (3) of section 4900 of the public health law, as amended by section 61 of chapter 219 of the laws of 2011 by to clarify that the definition of "emergency condition" applies regardless of the final diagnosis.
Section 8 amends paragraph (h) of subdivision (1) of section 4902 of the public health law, as added by section 10 of chapter 705 of the laws of 1996 to clarify that the definition of "emergency condition" applies regardless of the final diagnosis.
Section 9 amends paragraph (8) of subsection (a) of section 4902 of the insurance law, as added by section 16 of chapter 705 of the laws of 1996 to clarify that emergency services medically necessary to stabilize or treat an emergency condition are not subject to denial based on retro- spective review.
Section 10 amends Subsection (e) of section 4903 of the insurance law by requiring enhanced disclosure and reporting requirements by health insurers when making an adverse determination.
Section 11 amends paragraph (a) of subsection (5) of section 4903 of the public health law by requiring enhanced disclosure and reporting requirements by health insurers when making an adverse determination.
Section 12 provides that this act shall take effect on the sixtieth day after it becomes law and authorizes the promulgation of any necessary rules and regulations immediately.
Justification
The use of AI-based algorithms has become increasingly widespread in health insurance utilization review, yet existing state law does not adequately regulate how these tools are developed, used, or monitored. Without safeguards, AI systems can replicate bias, rely on incomplete data, and undermine clinical judgment, placing patient health and safety at risk.
AI tools can generate health coverage determinations, including denials, in a matter of seconds, often without meaningful review by a qualified clinical peer. This practice creates burdensome, costly, and time-con- suming barriers for providers seeking prior authorization to treat patients, delaying or denying medically necessary care.
This bill would establish clear consumer protections by requiring health insurers to disclose the use of AI algorithms in utilization review, mandating clinical peer review of adverse determinations and authorizing the Department of Financial Services to audit AI systems and enforce compliance through fines and license revocation. Together, these provision would bring much-needed transparency and accountability to utilization review processes, protect patients and providers, and safe- guard public health as the use of Al in health care continues to expand.
Legislative history
New bill
Fiscal implications
None.
Effective date
60 days after it shall have become law.
Summaries, headlines and scores are AI-generated from the official bill text. How we score bills. Source: NYS Open Legislation.