AI Healthcare Regulations in New Jersey (2025–2026)

New Jersey has no enacted AI-in-healthcare statute as of early 2026. Federal rules, including HIPAA, FDA SaMD guidance, and CMS conditions of participation, are the binding compliance layer. Several bills in the 222nd Legislature are in committee and could change this.

Quick Answer: Where New Jersey AI Healthcare Law Stands Now

The current requirement for New Jersey healthcare providers is to comply with federal law while monitoring state legislative developments.

New Jersey has not enacted a comprehensive AI-in-healthcare statute. As of March 2026, every AI-specific bill in the 222nd Legislature (2024–2026) remains in committee. The enacted distressed hospitals bill, S 2951 (P.L.2025, c.313), contains no AI-specific provisions.

Three federal frameworks are fully operative and enforceable:

• HIPAA Privacy and Security Rules (45 CFR Parts 160 and 164) apply to any AI system that touches protected health information, including ambient documentation tools, predictive analytics platforms, and AI-assisted diagnostics.
• FDA AI/ML-based Software as a Medical Device (SaMD) guidance, including the Predetermined Change Control Plan framework, governs AI tools that meet the definition of a medical device, such as AI-assisted radiology reads and sepsis prediction algorithms.
• CMS Conditions of Participation (42 CFR Part 482 for hospitals, 42 CFR Part 483 for skilled nursing facilities) require that AI-assisted care decisions do not violate patient rights or established care standards.

On the state side, the most direct healthcare-specific action is A 1160 (222nd Legislature), which directs the NJ Department of Health to evaluate technology uses in long-term care settings and clarifies that telehealth reimbursement parity covers long-term care. The bill has been referred to the Assembly Health Infrastructure Committee.

Build your compliance program on federal rules now. Assign a team member to track the four active NJ bills described below. Do not wait for a state statute to begin auditing your AI tool inventory.

Active New Jersey Bills That Affect AI in Healthcare

All four bills below were introduced in the 222nd Legislature and remain in their initial committee referrals as of March 18, 2026.

A 1160 (222nd Legislature) — Long-Term Care Technology and Telehealth Parity

What it does: Requires the NJ Department of Health to evaluate technology uses in long-term care settings, implements certain technological requirements for long-term care facilities, and clarifies that existing telehealth reimbursement parity covers long-term care settings.

Status: Introduced, referred to Assembly Health Infrastructure Committee (OpenStates, updated 2026-03-18).

Why it matters: This is the most direct healthcare-specific AI bill in the current session. A DOH evaluation mandate, if enacted, typically produces formal rulemaking, agency guidance, or procurement standards that facilities must meet to pass licensure inspections. Long-term care operators should begin documenting their current technology stack.

S 1840 (222nd Legislature) — New Jersey Responsible AI Advancement and Workforce Protection Act

What it does: Creates broad AI accountability requirements with a workforce protection framing. The bill's scope could reach AI-assisted clinical decision support tools and staffing algorithms used by NJ hospitals and health systems.

Status: Introduced in the Senate, referred to Senate Labor Committee (OpenStates, updated 2026-03-18).

Why it matters: If enacted, hospitals using AI for nurse staffing optimization, scheduling, or clinical decision support that affects workforce deployment would need to assess compliance. The workforce framing also raises questions about disclosure obligations to clinical staff when AI tools influence their assignments or workload.

S 1668 (222nd Legislature) — Artificial Intelligence Innovation Partnership

What it does: Establishes a funding mechanism for nonprofit partnerships that promote emerging AI technology businesses, including health-tech companies.

Status: Introduced in the Senate, referred to Senate Economic Growth Committee (OpenStates, updated 2026-03-18).

Why it matters: This is an opportunity bill, not a compliance burden. Health-tech startups and AI vendors in NJ should monitor it for grant eligibility if the bill advances.

A 2710 (222nd Legislature) — Authentic Relationships Act

What it does: Prohibits AI relationship simulation. The bill's scope has direct implications for AI-driven mental health chatbots, companion apps marketed to patients, and therapy-support tools that simulate therapeutic relationships.

Status: Introduced, referred to Assembly Science, Innovation and Technology Committee (OpenStates, updated 2026-03-18).

Why it matters: NJ behavioral health providers and digital health companies offering AI companion or therapy-support products to NJ residents should treat this bill as a near-term compliance risk. The definition of "relationship simulation" will be the critical threshold question.

What Changed Recently: 2025–2026 Legislative Activity

The 222nd Legislature introduced multiple AI-focused bills, a shift from prior sessions.

In the 217th, 218th, and 219th Legislatures, technology-related bills focused on R&D funding. The Edison Innovation Science and Technology Fund bills (S 3507/217, A 5123/217, A 1930/218, S 1921/218, A 2172/219) were about grant funding for university research, not AI regulation.

The 222nd Legislature's bills show regulatory intent. A 1160 creates a DOH mandate to evaluate and potentially regulate technology in care settings. S 1840 frames AI as a workforce issue requiring accountability. A 2710 proposes an outright prohibition on a specific AI use case.

DOH Evaluation Process: When the NJ Legislature directs DOH to evaluate a practice area, the agency generally conducts stakeholder hearings and then issues proposed rulemaking, informal guidance, or updated survey protocols.

Federal Context: The HHS Office for Civil Rights enforces HIPAA's application to AI systems that handle protected health information. Consult OCR for current guidance on this topic. Separately, the FDA continues to update its AI/ML-Based Software as a Medical Device Action Plan, including the Predetermined Change Control Plan framework for adaptive AI tools. Consult the FDA's Digital Health Center of Excellence for current documents.

Executive Order 14110: This October 2023 order directed HHS to establish an AI Task Force and produce guidance on AI safety in healthcare. Consult HHS directly for the current status of actions taken under this order.

Federal AI Healthcare Rules That Apply to New Jersey Providers Today

HIPAA Privacy and Security Rules

Any AI system that processes, stores, or transmits protected health information must meet HIPAA safeguards (45 CFR Parts 160 and 164). This includes ambient documentation tools, predictive analytics platforms, and AI diagnostic tools. Business associate agreements must address AI-specific data handling. Consult HHS for guidance on how HIPAA's minimum necessary standard applies to training data sets derived from PHI.

FDA Software as a Medical Device Regulation

AI tools that meet the FDA's definition of a medical device require clearance or approval before clinical use (FDA 21 CFR Part 820; SaMD guidance). Examples include AI-assisted radiology reads and sepsis prediction algorithms. Tools that qualify as non-device clinical decision support software under 21st Century Cures Act Section 3060 are exempt, but the exemption has specific criteria. Consult the FDA's "Clinical Decision Support Software" guidance (September 2022) for the framework distinguishing regulated SaMD from exempt CDS software.

21st Century Cures Act Information Blocking

AI-generated clinical data, including AI-produced diagnostic impressions and predictive risk scores incorporated into the medical record, may not be withheld in ways that constitute information blocking (21st Century Cures Act, Pub. L. 114-255; 45 CFR Part 171).

FTC Act Section 5

Unfair or deceptive claims about AI tool accuracy, capabilities, or clinical validation in healthcare marketing are actionable under the FTC Act (15 U.S.C. § 45). This applies to AI vendors marketing to NJ health systems and to providers making claims to patients about AI-assisted care.

CMS Conditions of Participation

Hospitals must ensure AI-assisted care decisions do not violate patient rights or established care standards (42 CFR Part 482). Skilled nursing facilities face the same requirement under 42 CFR Part 483.

NJ Consumer Fraud Act

The NJ Consumer Fraud Act (N.J.S.A. 56:8-1 et seq.) applies to deceptive marketing of AI health products to NJ consumers. AI vendors making unsupported efficacy claims face exposure under this statute in addition to FTC Act liability.

Compliance Requirements by Healthcare Setting: Comparison Table

All bill-based requirements listed below are pending enactment as of March 2026. Federal rules in the rightmost column are currently enforceable.

Key Definitions and Scope Questions Under NJ's Proposed AI Rules

The key compliance question for any proposed rule is its scope. The definitions below are based on current bill language and federal frameworks; definitions in New Jersey bills may change during the legislative process.

S 1840: What Counts as "Artificial Intelligence"?

The bill's definition section should be reviewed against the enrolled text, which may be updated as the bill moves through committee. Consult the NJ Legislature website at njleg.state.nj.us for current bill text. The threshold question for hospitals is whether clinical decision support software, predictive analytics, and ambient documentation tools fall within the bill's AI definition. Broad definitions would likely capture all three.

A 1160: Is "Technology Uses" Limited to AI?

A 1160's mandate covers "technology uses in long-term care settings" broadly. Based on the bill's description (NJ A 1160, 222nd Legislature), the scope extends beyond AI to include robotics, sensor systems, remote monitoring devices, and EHR automation. Long-term care operators should not assume the DOH evaluation will be limited to machine learning tools.

A 2710: Where Does "Relationship Simulation" End and Legitimate Therapy Support Begin?

A 2710 (222nd Legislature) prohibits "artificial intelligence relationship simulation," but the boundary between a prohibited simulation and a legitimate AI-assisted therapy support tool is not defined in the bill's title. Consult the current bill text at njleg.state.nj.us. Behavioral health providers using AI tools that maintain conversational continuity with patients, simulate emotional connection, or are marketed as companions should treat this bill as a direct compliance risk until the definition is clarified.

FDA SaMD vs. Exempt Clinical Decision Support

The FDA's "Clinical Decision Support Software" guidance (September 2022) is the operative framework for distinguishing regulated SaMD from CDS software exempt under 21st Century Cures Act Section 3060. The four-part test for CDS exemption requires that the software: (1) not be intended to acquire, process, or analyze medical images or signals; (2) display, analyze, or print medical information; (3) support or provide recommendations to a clinician; and (4) enable the clinician to independently review the basis for the recommendation. AI tools whose reasoning is not reviewable by the clinician are more likely to be regulated SaMD.

Next Steps and Who to Contact in New Jersey

Five Compliance Steps to Take Now

Step 1: Inventory all AI tools currently in use. Categorize each tool by function (clinical vs. administrative), whether it touches PHI, and whether it is patient-facing.

Step 2: Assess each tool against current binding rules. Run each clinical AI tool through the FDA SaMD/CDS exemption analysis. Confirm HIPAA business associate agreements are in place for all AI vendors handling PHI. Check CMS CoP compliance for any AI tool affecting care decisions.

Step 3: Assign bill-tracking responsibility. Designate a staff member or counsel to monitor A 1160, S 1840, A 2710, and S 1668. Set up alerts at njleg.state.nj.us.

Step 4: Long-term care facilities should prepare for DOH engagement. If A 1160 advances, the DOH evaluation will likely include a stakeholder comment period. Facilities with a documented technology inventory will be better positioned to participate.

Step 5: Health-tech companies should monitor S 1668 for funding opportunities. If the Artificial Intelligence Innovation Partnership is enacted, grant eligibility criteria will be published by the administering entity.

Contact Information

NJ Department of Health The Office of Health Care Quality Assessment is the office most likely to administer any technology evaluation mandate from A 1160. Verify current contact information at nj.gov/health.

NJ Legislature Bill Tracking Search by bill number at njleg.state.nj.us for real-time status, committee schedules, and current bill text.

NJ Assembly and Senate Committees For stakeholder input on pending bills, contact the relevant committee through the NJ Legislature's committee directory at njleg.state.nj.us.

• A 1160: Assembly Health Infrastructure Committee
• S 1840: Senate Labor Committee

FDA Digital Health Center of Excellence For SaMD regulatory pathway questions: digitalhealth@FDA.hhs.gov.

HHS Office for Civil Rights For questions about HIPAA's application to AI systems: 1-800-368-1019 or through hhs.gov/ocr.

Outside Counsel This page provides a regulatory framework, not legal advice. Facility-specific compliance analysis and vendor contract review require a NJ-licensed healthcare attorney.

Page reflects legislative status as of March 2026 based on OpenStates records updated 2026-03-18. Bill status and federal guidance documents change frequently. Verify current status of all cited bills at njleg.state.nj.us and confirm current federal guidance versions with the relevant agencies before making compliance decisions.