Pennsylvania AI Laws and Regulation (2026)

Pennsylvania regulates AI through two enacted criminal laws: Act 125 of 2024 extends the state's sexual-image offenses to AI-generated deepfake material, and Act 35 of 2025 makes it a misdemeanor to distribute an AI forgery of a real person as genuine with intent to defraud or injure, rising to a felony when that is done through a scheme to defraud, coerce or commit theft. Executive Order 2023-19 governs AI use within state agencies.

Pennsylvania has taken a multi-pronged approach to artificial intelligence regulation, combining enacted criminal laws targeting deepfakes and digital forgery with aggressive executive-branch AI governance and a growing legislative pipeline. Governor Josh Shapiro has positioned the Commonwealth as a national leader in government AI adoption while simultaneously signing laws to protect residents from AI-generated harms.

This article covers every enacted Pennsylvania AI law, pending legislation, executive policies, and federal considerations as of March 2026.

This article is for informational purposes only and does not constitute legal advice. Consult a Pennsylvania-licensed attorney for guidance on specific situations.

Act 125 of 2024: AI-Generated Deepfake Images (Enacted)

Pennsylvania's first enacted AI law is Act 125 of 2024, derived from Senate Bill 1213. The bipartisan legislation, sponsored by Sen. Tracy Pennycuick (R-24), Sen. Jimmy Dillon (D-5), and Sen. Lisa Boscola (D-18), passed the Senate unanimously in June 2024 and was signed into law in October 2024.

What the Law Addresses

Act 125 amends Titles 18 (Crimes and Offenses) and 61 (Prisons and Parole) of the Pennsylvania Consolidated Statutes to address the use of AI to generate deepfake child sexual abuse material. The AI substance is all in Title 18. The Title 61 amendments in Sections 8 through 10 of the act are parole-eligibility conforming changes that follow from renaming child pornography as child sexual abuse material throughout the code, not AI provisions. Prior to this law, Pennsylvania made it illegal to share intimate images without consent but did not clearly address AI-generated content.

Key Provisions

Section 3 of the act rewrote 18 Pa.C.S. 6312(c) and (d), and Section 2 rewrote 18 Pa.C.S. 3131, so that the following now reach artificially generated material:

• Selling, distributing, delivering, disseminating, transferring or displaying artificially generated child sexual abuse material, or possessing it for any of those purposes, under 18 Pa.C.S. 6312(c)
• Intentionally viewing, or knowingly possessing or controlling, artificially generated child sexual abuse material, under 18 Pa.C.S. 6312(d)
• Disseminating an artificially generated sexual depiction of an individual with intent to harass, annoy or alarm them, under 18 Pa.C.S. 3131(a)(2)

Act 125 did not amend 18 Pa.C.S. 6312(b), the provision covering photographing or causing a child to engage in a prohibited sexual act. Generating the material is therefore not itself the added offense, and is reached only through the possession, control and distribution prongs above.

Penalties

Disseminating an artificially generated sexual depiction under 18 Pa.C.S. 3131 is graded a first-degree misdemeanor when the person depicted is a minor and a second-degree misdemeanor when the person depicted is not. The law provides law enforcement with the authority to prosecute individuals who distribute or possess AI-generated child sexual abuse materials that were previously difficult to prosecute under existing statutes.

Act 35 of 2025: Digital Forgery Law (Enacted)

On July 7, 2025, Governor Josh Shapiro signed Act 35 of 2025 (formerly Senate Bill 649), establishing criminal penalties for distributing deepfakes with fraudulent or injurious intent. The act was not effective on signing. Section 2 of the enrolled act reads that it shall take effect in 60 days, so 18 Pa.C.S. 4101.1 has applied since September 5, 2025.

What the Law Criminalizes

Act 35 adds 18 Pa.C.S. 4101.1, the offense of digital forgery, to Pennsylvania's forgery and fraudulent practices chapter.

The offense has a narrower reach than the phrase "AI impersonation" suggests. Under Section 4101.1(a), a person is guilty only if, with intent to defraud or injure anyone, or with knowledge and intent that the person is facilitating a fraud or injury to be perpetrated by anyone, the person both generates or creates and distributes a forged digital likeness as genuine, and knows or reasonably should know that the visual representation or audio recording is a forged digital likeness.

Both halves matter. A nonconsensual deepfake or voice clone that is created but never distributed is not an offense under this section, and neither is one distributed without fraudulent or injurious intent. The law targets people who use AI to defraud or harm Pennsylvanians.

Penalty Structure

Intent to defraud or injure is an element of every violation, so it is not what raises the grade. Under Section 4101.1(b)(2), the third-degree felony applies only when the violation is committed through involvement in a scheme to defraud, coerce or commit theft of monetary assets or property, including schemes like faking a relative's voice to trick older adults into sending money.

Exceptions and Defenses

The law provides exceptions for:

• Constitutionally protected activity (satire, parody, commentary)
• A law enforcement officer engaged in the performance of official duties
• A provider or developer of any technology used in the creation of a visual representation or audio recording
• Any information service, system or access software provider that provides or enables computer access by multiple users to a computer server, including specifically a service or system that provides access to the internet

The last two exceptions are written without conditions. Subsection (c) attaches no requirement that a technology provider, developer or internet service have avoided facilitating the forgery.

An affirmative defense exists when the defendant took reasonable action to notify viewers or listeners that the forged digital likeness was not genuine.

Executive Order 2023-19: Government AI Governance

Governor Shapiro signed Executive Order 2023-19 in September 2023, establishing the framework for AI use within Pennsylvania state government. The order set clear principles for generative AI use: accuracy, transparency, security, fairness, and employee empowerment.

Generative AI Governing Board

The executive order created the Generative AI Governing Board to oversee AI policy and implementation across all Commonwealth agencies. The board is responsible for reviewing AI deployments, setting standards, and ensuring ethical use.

OpenAI Pilot Program

In January 2024, Pennsylvania launched a first-in-the-nation pilot program with OpenAI, providing state employees access to ChatGPT for work tasks. The pilot results showed employees reported an average time savings of 95 minutes per day using AI for writing, research, summarization, and IT support.

University of Pennsylvania Partnership

In October 2025, the Shapiro Administration announced a Cooperative Agreement for AI Advising Services with the University of Pennsylvania to advance responsible AI use across state government.

National Recognition

Under Governor Shapiro, Pennsylvania became one of the top three states in the nation for AI readiness, according to the July 2025 Code for America Government AI Landscape Assessment.

Commonwealth AI Policy (January 2026)

Pennsylvania's Office of Administration published an updated Artificial Intelligence Policy effective January 13, 2026, replacing the previous August 2025 version. The policy applies to all offices, departments, boards, commissions, and councils under the Governor's jurisdiction.

Key Policy Requirements

The AI policy establishes guidelines for integrating AI into Commonwealth operations. Notable provisions include:

• No final employment decisions by AI: The policy prohibits using generative AI to make final decisions that affect employment
• Data protection: Sensitive government data requires approved AI systems with appropriate security controls
• Transparency: Agencies must document AI use in decision-making processes
• Human oversight: AI outputs must be reviewed by qualified personnel before being used in consequential decisions

Joint State Government Commission AI Report (January 2026)

In January 2026, Pennsylvania's Joint State Government Commission released a 389-page report titled "Artificial Intelligence: Advisory Committee Recommendations on the Adoption and Use of AI in Pennsylvania." The report, mandated by House Resolution 170, provides a comprehensive assessment of AI's impact across healthcare, education, government, and law enforcement.

Key Recommendations

The report's recommendations include:

Government Oversight

• Create a Chief Artificial Intelligence Officer position for the Commonwealth
• Review AI legislation every three to five years
• Require AI developers to assess systems' potential impact on privacy, due process, and equal protection

Business and Employment

• Require employers and businesses to disclose when and how AI is used
• Communicate when automated decision-making tools make "high-risk decisions" such as hiring, college admissions, and healthcare decisions
• Give employees "reasonable advance notice" before deploying AI tools that could affect their employment
• Require employers to report when AI displaces employees

Education

• AI should only be implemented when empirical evidence supports student benefit
• AI tools should be available to all students equitably
• Teachers and educators should receive training on ethical AI responsibilities

Political Advertising

• Require disclosure when AI is used for political and election-related purposes

SAFECHAT Act: AI Chatbot Child Safety (SB 1090, 2026 Pending)

The Safeguarding Adolescents from Exploitative Chatbots and Harmful AI Technology Act (SAFECHAT Act, SB 1090) passed the Pennsylvania Senate 49-1 on March 17, 2026. The bill was sponsored by Senators Pennycuick and Miller.

Key Requirements

The SAFECHAT Act would require AI companion operators to:

• Issue "clear and conspicuous" notification that an AI companion is artificially generated
• Establish protocols to prevent chatbots from producing content related to suicide, self-harm, or violence
• Direct users to crisis resources when high-risk language is detected
• Implement age-appropriate standards for chatbots interacting with minors

Penalties

Current Status

The SAFECHAT Act now awaits consideration in the Pennsylvania House of Representatives.

AI in Healthcare: HB 1925 (Pending)

House Bill 1925 proposes new regulations for AI use in healthcare and insurance. Introduced on October 6, 2025, the bill would amend the Health and Safety and Insurance titles of Pennsylvania law.

Key Provisions

HB 1925 would require:

• Healthcare providers, insurers, and managed care plans to disclose when AI is used in clinical decision-making or utilization review
• AI tools cannot override human judgment; a qualified professional must make the final determination
• Insurers must reveal when AI drives a denial or prior authorization decision
• Facilities and insurers must submit compliance information to the Department of Health or Insurance

Current Status

HB 1925 remains in the House Communications and Technology Committee, where a public hearing was held on the proposal.

AI in Elections (Pending)

Pennsylvania does not yet have an enacted law regulating AI in elections, although the House has passed legislation addressing the issue.

House Deepfake Disclosure Bill

The Pennsylvania House unanimously passed a bill requiring campaigns to disclose when they use AI-generated deepfakes that mimic a candidate's appearance or voice. Campaigns that fail to disclose could be fined for every day the ad runs. Candidates would be able to sue other candidates, political parties, or PACs for "knowingly and intentionally" disseminating undisclosed deepfakes within 90 days of an election.

The bill currently sits in the state Senate Communications and Technology Committee.

Senate Proposals

Senator Lindsey M. Williams (D-Allegheny) announced on March 17, 2026 that she will introduce separate legislation requiring political advertisements to disclose if AI was used in their creation.

AI Data Centers and Economic Development (SB 939, Pending)

Senate Bill 939, sponsored by Sen. Greg Rothman (R), aims to position Pennsylvania as a hub for AI infrastructure. The bill would establish an Office of Transformation and Opportunity and create an AI, Data Center, and Emerging Technology Regulatory Sandbox Program.

Regulatory Sandbox

The sandbox program would allow companies to test innovative AI products or services for up to 12 months without obtaining a license that might otherwise be required, under government oversight.

SB 939 was amended and approved by the Senate Committee on Communications and Technology in February 2026 with a unanimous 10-0 vote.

AI in Employment

Pennsylvania has not enacted AI-specific employment legislation. However, the Commonwealth's January 2026 AI Policy prohibits state agencies from using generative AI to make final employment decisions. The Joint State Government Commission's report recommends extending similar requirements to the private sector.

Existing anti-discrimination laws under the Pennsylvania Human Relations Act apply to AI-powered employment tools. Employers using automated systems in hiring must ensure compliance with state and federal equal employment opportunity requirements.

The state legislature has also considered proposals requiring the Department of Labor and Industry to study AI's future impact on the labor force.

Federal AI Policy and Pennsylvania

Federal AI policy intersects with Pennsylvania's regulatory efforts in several ways.

Executive Order 14365 (December 2025)

President Trump's Executive Order 14365 aims to preempt state AI laws that "unduly impede" innovation. This creates potential tension with Pennsylvania's pending AI healthcare regulation (HB 1925) and SAFECHAT Act.

However, the order's carve-outs for child safety measures could protect the SAFECHAT Act. Pennsylvania's enacted deepfake laws (Act 125 and Act 35) address criminal conduct rather than AI development, making them less susceptible to preemption challenges.

Governor Shapiro's AI Approach

Governor Shapiro has positioned Pennsylvania as both pro-innovation and pro-protection, creating an AI Task Force to address harmful AI chatbots while maintaining the Commonwealth's AI-friendly business environment. This balanced approach may serve as a model for navigating federal preemption pressures.

Summary of Pennsylvania AI Laws and Actions

More Pennsylvania Laws

Explore other Pennsylvania law topics on Recording Law:

• Pennsylvania Recording Laws
• Pennsylvania Data Privacy Laws
• Pennsylvania Biometric Privacy Laws
• Pennsylvania Whistleblower Laws
• Pennsylvania Sexting Laws
• Pennsylvania Background Check Laws
• Pennsylvania Surveillance Camera Laws