Oklahoma AI Laws and Regulation (2026)

Oklahoma regulates AI primarily through targeted legislation rather than a single comprehensive statute. HB 1364, enacted in May 2025 and effective November 1, 2025, criminalizes nonconsensual AI-generated intimate images. Executive Order 2023-24 established AI governance standards for state agencies, and several broader bills are pending in the 2025 and 2026 sessions.

Oklahoma does not yet have a single comprehensive artificial intelligence statute. However, the state has enacted targeted AI legislation, issued executive-level AI governance standards, and introduced sweeping proposals in the 2025 and 2026 legislative sessions. The state's approach blends executive action with incremental legislative steps, addressing deepfakes, child safety, healthcare AI, government operations, and workforce development.

This article covers every enacted Oklahoma AI law, pending legislation, executive policies, and federal preemption considerations as of September 2026.

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

Oklahoma's AI Executive Order and State Government Standards

Governor J. Kevin Stitt took the first major step toward AI governance with Executive Order 2023-24, signed on September 25, 2023. The order created the Governor's Task Force on Emerging Technologies: Artificial Intelligence Strategy to Support State Agencies in Oklahoma.

The task force, chaired by State CIO Joe McIntosh, focused on four strategic goals: economic enhancement, ethical AI implementation, educational and professional development, and fostering innovation within state government.

The OMES AI Standard

Following the task force's recommendations, the Office of Management and Enterprise Services (OMES) published the official Use of AI in Oklahoma State Government Standard. This document governs how all executive-branch agencies acquire, deploy, and manage AI systems.

Key requirements of the OMES standard include:

• Security reviews: All AI systems must complete a third-party security review approved by the OMES Chief Information Security Officer before deployment.
• Sensitive data protections: HIPAA, FERPA, FTI, PII, and CJIS data cannot be transmitted through public AI systems. A separate secure instance requires CIO approval.
• Transparency and accountability: State agencies must maintain records of AI system decisions and ensure human oversight of consequential determinations.
• Employee compliance: Violations of the standard may result in disciplinary action, up to and including termination.

All state agencies, boards, and commissions under the CIO's authority are required to adhere to the most current published standard.

HB 1364: AI-Generated Revenge Porn Law (Enacted 2025)

Oklahoma's most significant enacted AI law is House Bill 1364, signed by Governor Stitt in May 2025. The law expands Oklahoma's existing revenge porn statute, 21 O.S. Section 1040.13b, to cover artificially generated sexual depictions.

What the Law Prohibits

Under HB 1364, it is a criminal offense to disseminate an artificially generated sexual depiction of another person without that person's effective consent, with the intent or with reckless disregard to harass, annoy, threaten, alarm, or cause physical, emotional, reputational, or economic harm to the depicted person. The offense is dissemination. Creating an image and keeping it to yourself is not itself covered by this section.

The law defines an "artificially generated sexual depiction" as a visual depiction that appears to authentically depict an individual in a state of nudity or engaged in sexual conduct that did not occur in reality, and whose production was substantially dependent on technical means, including artificial intelligence or photo-editing software, rather than on another person's ability to physically impersonate the depicted person.

Penalties

HB 1364 does not carry a single penalty. The statute tiers the offense, and the misdemeanor is only the floor:

Subsection H removes the state's discretion to file a subsection G violation as a misdemeanor, so the commercial and extortion-adjacent conduct the statute targets cannot be charged down. A person convicted under the section is also subject to the forfeiture provisions of Section 1040.54, and the court may order the defendant to remove the disseminated material where it is within the defendant's power to do so.

Exceptions

The law provides exceptions for lawful uses, including dissemination for a lawful criminal investigation, dissemination in connection with reporting unlawful conduct, images or depictions involving voluntary exposure in public or commercial settings, and dissemination that serves a lawful purpose. Interactive computer services, wireless service providers, and telecommunications or broadband providers are shielded from liability based solely on content provided by another person.

Effective Date

HB 1364 took effect on November 1, 2025. With its enactment, Oklahoma joined a growing majority of states that explicitly prohibit nonconsensual AI-generated intimate imagery.

Two Competing Versions of Section 1040.13b

Oklahoma's official Title 21 codification currently publishes Section 1040.13b in two versions from the same 2025 session, and readers should know which one they are looking at.

Version 1 is the HB 1364 text (Laws 2025, c. 23, Section 1, effective November 1, 2025). It contains the artificial intelligence definitions, the artificially generated sexual depiction offense, and the penalty figures described above.

Version 2 is the HB 2104 text (Laws 2025, c. 486, Section 399, effective January 1, 2026), enacted as part of Oklahoma's felony reclassification overhaul. It omits every artificial intelligence definition and the AI-specific offense entirely, and it restates the subsection G and K felonies as a Class D1 felony offense punishable as provided in subsections B through F of Section 20N of Title 21. Under Section 20N, a Class D1 offense carries up to five years in the custody of the Department of Corrections, with at least twenty percent of the sentence served before release; a defendant with prior qualifying convictions faces one to seven years, or two to ten years.

The published statutes do not reconcile the two versions. Anyone relying on Section 1040.13b for conduct on or after January 1, 2026 should confirm with an Oklahoma attorney which version controls, because the AI-specific offense appears only in version 1.

HB 3299: Comprehensive Deepfake Criminalization (2026 Pending)

House Bill 3299, authored by Rep. Neil Hays (R-Checotah), would significantly expand Oklahoma's deepfake protections beyond intimate imagery. The bill passed the House Criminal Judiciary Committee unanimously in February 2026 and cleared the House Judiciary and Public Safety Oversight Committee on February 26, 2026. No further legislative action is recorded as of September 2026.

Scope of the Bill

HB 3299 would make it unlawful to create and distribute digitized or synthetic media depicting another person's name, image, voice, or likeness without written consent when done with intent to cause emotional, financial, reputational, or physical harm.

Tiered Criminal Penalties

The bill establishes two tiers of criminal penalties:

Advertising Disclosure Requirement

HB 3299 also includes a provision for political advertising. Media advertising agencies would be required to obtain a signed attestation from content creators stating whether the advertisement contains digitized or synthetically altered material. Any digital or synthetic media in political advertising would need to be disclosed.

If enacted, the bill would position Oklahoma alongside Tennessee, which passed the ELVIS Act in 2024, in providing broad protections for individuals against unauthorized use of their likeness through AI.

SB 746: Political Deepfake Disclosure (2025 Pending)

Senate Bill 746, authored by Sen. Ally Seifried (R-Claremore), addresses AI-generated content specifically in the context of political campaigns.

Key Requirements

Under SB 746, candidates, political parties, and committees would be required to disclose any political advertisement that uses generative AI to portray real people in fabricated situations. The disclosure must be clear and visible in images and video, or audible in audio-only content.

Legal Remedies for Candidates

The bill would grant candidates new legal tools to combat AI misuse in elections, including the ability to sue to stop distribution of manipulated content or seek monetary damages when their likeness, voice, or actions are misrepresented.

Sen. Seifried has explained that the bill extends the same transparency requirements that already apply to campaign funding disclosures: "Campaign ads and mailers already have to disclose who paid for them, and SB 746 applies that same level of transparency to content created or altered using generative AI."

The bill was referred to the Senate Technology and Telecommunications Committee on February 4, 2025 and has not received a recorded vote in either chamber as of September 2026.

SB 1521: AI Chatbot Protections for Minors (2026, Passed Legislature)

Senate Bill 1521, authored by State Sen. Warren Hamilton, targets AI chatbots that may endanger children. The bill passed the Senate Technology and Telecommunications Committee 8-0 on February 19, 2026, then passed the full Senate 43-0 on March 23, 2026, and the full House 90-0 on April 28, 2026. The most recent recorded legislative action is its engrossment back to the Senate on April 29, 2026 for concurrence on House amendments. It has not reached the Governor's desk.

Prohibited Chatbot Conduct

The bill would ban AI chatbots that encourage minors to engage in:

• Sexually explicit content
• Suicide or self-harm
• Violence

Age Verification Requirements

SB 1521 requires AI developers to implement age verification by validating a state-issued ID, rather than relying on self-reported age. Developers must also disclose to users that they are interacting with AI and freeze accounts until age is verified.

Penalties

Violations could bring civil penalties of up to $100,000 per incident. The legislation was motivated by reported cases of teen deaths linked to AI chatbot interactions.

HB 1916: The Responsible Deployment of AI Systems Act (2025 Pending)

House Bill 1916 represents Oklahoma's most ambitious AI proposal. Introduced on January 16, 2025, the Responsible Deployment of AI Systems Act would establish a comprehensive regulatory framework for artificial intelligence in the state.

Four-Tier Risk Classification

The bill classifies AI systems into four risk categories:

Prohibited AI Systems

Under HB 1916, the following AI uses would be banned as "unacceptable risk":

• AI systems used for social scoring
• Manipulative AI targeting vulnerable groups
• Unregulated real-time biometric surveillance
• AI-driven misinformation campaigns targeting elections, public health, or emergencies

New State Bodies and Programs

The bill proposes three new entities:

1. AI Council: An oversight body to monitor AI deployment across the state and guide regulatory policy.
2. AI Regulatory Sandbox Program: A testing environment for innovative AI technologies to operate under relaxed regulations with government oversight.
3. Workforce Development Program: A program focused on supporting small businesses and underserved communities in AI integration and training.

As of September 2026, HB 1916 had recorded no action since it was referred to the Rules Committee following its second reading on February 4, 2025.

AI in Healthcare: HB 3577 (2024, Did Not Pass)

Oklahoma considered but did not enact HB 3577, the Artificial Intelligence Utilization Review Act (AURA), during the 2024 session. The bill would have regulated the use of AI algorithms in health insurance utilization review processes.

Key Provisions (Not Enacted)

HB 3577 would have required:

• Healthcare insurers to disclose any use of AI algorithms in utilization review on their websites
• Submission of algorithms and training datasets to the Oklahoma Insurance Department for transparency review
• Attestation of bias minimization in AI systems
• Human review of specialist denial decisions, with documentation of individual clinical records
• Civil penalties for violations, with caps on fines

The bill passed the House Government Modernization and Technology Committee but died before reaching the full legislature. Its provisions may resurface in future sessions as other states advance similar healthcare AI transparency requirements.

AI in State Government Operations

Beyond the OMES standard, Oklahoma has begun integrating AI into actual government operations. The state has deployed AI in its procurement system through a partnership with Celonis, using process-mining AI to identify filing errors and streamline purchasing.

HB 3828: State AI Inventory (2024, Did Not Pass)

HB 3828 would have required OMES and the Administrative Office of the Courts to inventory all AI systems by December 31, 2024, and annually thereafter. State agencies outside OMES would have been required to conduct their own inventories and post results publicly online.

The bill also proposed assessments of agency AI systems to ensure they do not discriminate. While the bill passed committee unanimously, it did not advance further during the 2024 session.

AI in Employment

Oklahoma has not enacted legislation specifically regulating AI in employment decisions. However, the state's existing employment and anti-discrimination laws apply to AI tools used in recruiting, scheduling, workforce analytics, and performance evaluation.

Employers using AI in hiring must ensure that automated tools do not replace human judgment entirely. The ACLU of Oklahoma has raised concerns about algorithmic bias in hiring systems, noting that AI-powered recruitment tools can perpetuate racial, gender, and disability discrimination if not properly audited.

Under HB 1916 (if enacted), AI tools used in high-stakes employment decisions such as hiring, promotion, and termination would likely fall under the "High Risk" classification, triggering strict oversight requirements.

Federal AI Policy and Oklahoma

Federal AI policy creates both opportunities and tensions for Oklahoma's regulatory approach.

Executive Order 14365 (December 2025)

President Trump's Executive Order 14365 directs the Attorney General to identify and challenge state AI laws that "unduly impede" AI innovation. The order signals a preference for minimal state-level AI regulation and threatens to preempt state laws that conflict with federal AI priorities.

However, the order includes carve-outs for child safety measures and state government AI procurement policies. These carve-outs could protect Oklahoma proposals like SB 1521 (chatbot protections for minors) and the OMES AI standard from federal preemption challenges.

Oklahoma Attorney General Response

Oklahoma's Attorney General signed a multi-state letter critical of congressional efforts to prohibit state AI regulation, asserting that states retain authority to protect their residents from AI-related harms.

This tension between federal preemption and state regulatory authority will shape Oklahoma's AI legislative agenda in the coming years.

Summary of Oklahoma AI Laws and Bills

More Oklahoma Laws

Explore other Oklahoma law topics on Recording Law:

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