Pennsylvania AI Laws and Regulation (2026)

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. Prior to this law, Pennsylvania made it illegal to share intimate images without consent but did not clearly address AI-generated content.

Key Provisions

The law makes it a crime to:

• Create AI-generated child sexual abuse material
• Harass someone by distributing a deepfake image depicting them in a state of nudity or engaged in a sexual act without their consent
• Possess or distribute AI-generated sexually explicit images of minors

Penalties

Offenses carry enhanced penalties when the victim is a minor. The law provides law enforcement with the authority to prosecute individuals who generate and disseminate 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 creating or disseminating deepfakes with fraudulent or injurious intent.

What the Law Criminalizes

Act 35 amends Pennsylvania's existing forgery provisions to include "forged digital likeness." The law creates criminal penalties for anyone who uses AI to produce nonconsensual forged digital likenesses, including deepfakes and voice clones, to defraud or harm Pennsylvanians.

Penalty Structure

Offense	Classification	Severity
Nonconsensual digital impersonation	First-degree misdemeanor	Up to 5 years imprisonment
Digital impersonation with fraudulent intent	Third-degree felony	Up to 7 years imprisonment

The third-degree felony classification applies when AI-generated content is used for fraud, 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)
• Official law enforcement activities
• Technology companies that provide the means to create deepfakes, as long as they did not intentionally facilitate creation and dissemination

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

Violation Type	Maximum Fine
Chatbot disclosure failures	$50,000 per violation
Content safeguard failures resulting in harm to a minor	$500,000 per incident

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

Law/Action	Year	Status	Topic	Key Provision
Exec. Order 2023-19	2023	Active	Government AI governance	Created Generative AI Governing Board
Act 125 (SB 1213)	2024	Enacted	AI-generated CSAM	Criminalizes AI deepfake child sexual abuse material
Act 35 (SB 649)	2025	Enacted (July 7, 2025)	Digital forgery	AI impersonation is misdemeanor; fraud is felony
Commonwealth AI Policy	2026	Active	Government AI standards	Prohibits AI from making final employment decisions
JSGC AI Report	2026	Advisory	Comprehensive AI recommendations	389-page report recommending Chief AI Officer
SB 1090 (SAFECHAT)	2026	Passed Senate 49-1	Child safety	AI chatbot safeguards, fines up to $500,000
HB 1925	2025	Pending in committee	Healthcare AI	Requires AI disclosure in clinical decisions
SB 939	2025	Pending	AI data centers	Regulatory sandbox for AI companies
Election deepfake bill	2025	Passed House, pending in Senate	Election AI	Requires disclosure of AI in campaign ads

More Pennsylvania Laws

Explore other Pennsylvania law topics on Recording Law:

• Pennsylvania Recording Laws
• [Pennsylvania Data Privacy Laws](/us-laws/data-privacy-laws/pennsylvania-data-privacy-laws)
• Pennsylvania Biometric Privacy Laws
• Pennsylvania Whistleblower Laws
• Pennsylvania Sexting Laws
• Pennsylvania Background Check Laws
• Pennsylvania Surveillance Camera Laws