South Carolina's AI policy posture is more developed than its national reputation reflects. The state has a published AI strategy, a Center of Excellence, a first-in-the-nation House committee with explicit AI legislative authority, an interim judicial-branch policy, and at least five AI-specific bills of the 126th General Assembly in committee, plus two AI-related resolutions.
SCAIO maintains a live policy tracker on the home page that updates as new items emerge. This chapter synthesizes the picture as of the publication of this edition. Readers wanting current bill activity should consult the live tracker; this chapter sets the structural context.
Compared to peer states, South Carolina has an unusually structured starting position on AI policy. Four institutional pieces are in place:
Several states have one or two of these. SC has all four. The picture is early-stage but coherent.
The South Carolina Department of Administration's AI Strategy is the state's foundational executive-branch document on AI. The strategy lays out principles for state-agency AI adoption — including responsible use, data governance, workforce considerations, and risk management — and designates the AI Center of Excellence as the coordinating body for cross-agency implementation. Coverage in HSToday at the time of release described the strategy as one of the more substantive state-level AI documents then published.
The strategy is best understood as a posture-setting document rather than a regulatory rulebook. It does not specify which AI tools agencies may or may not procure, does not specify procurement-evaluation rubrics, and does not establish enforcement mechanisms with teeth. It does establish an expectation that state agency AI deployment will be coordinated, surveyable, and aligned to a shared set of principles. Whether that expectation translates to operational reality across SC's 60-plus state agencies is one of the more important things to watch over the next 18 months.
The AI Center of Excellence is the operational arm of the state's executive-branch AI posture. The Center's role is to:
As of this edition, the Center of Excellence does not yet have a dedicated public-facing website. The SCAIO ecosystem directory currently links to GovTech's coverage of the Center's establishment as an interim resource. The absence of a Center-of-Excellence public web presence is a gap worth closing — both for transparency and as a coordination tool. (The framing in this report is constructive: a CoE website would help the work the CoE is already doing become more visible and more accessible to the SC institutions and citizens it is intended to serve.)
The 126th General Assembly (2025–2026) has at least five AI-specific bills in committee, plus two AI-related resolutions — substantially more legislative activity than is generally appreciated. The two most-developed bills are H.5253 and S.443; three additional bills (S.963, S.1037, H.5138) cover algorithmic-discrimination consumer protection and chatbot regulation. Each substantive bill is summarized below; the AI-related resolutions are noted at the end of the section.
Sponsored by Reps. Brandon Guffey (R-48) and Donald G. "Don" Chapman (R-8), H.5253 was introduced February 24, 2026 and referred to the House Committee on Education and Public Works. The bill adds Section 59-28-195 to the South Carolina Code, establishing limitations and safeguards on AI use in K–12 public schools. The provisions are detailed — significantly more so than peer-state AI-in-education bills:
H.5253 is one of the more comprehensive state AI-in-education bills introduced in any state in 2025–2026. The opt-in model (rather than the opt-out model favored in some peer-state proposals) is a substantive choice with implementation implications for school districts and ed-tech vendors. Whether the bill advances out of committee in this General Assembly session is, at the time of writing, an open question.
Sponsored by Sen. Sabb, S.443 was introduced March 11, 2025 and referred to the Senate Committee on Banking and Insurance. The bill adds Section 38-59-23 to the South Carolina Code (Insurance), targeting one of the most contested AI use cases in U.S. healthcare: insurer use of automated decision-making tools to deny or modify prior-authorization requests.
Substantively, S.443:
The bill is structurally similar to California's SB 1120 (the "Physicians Make Decisions Act") and to bills introduced in several other states. It targets the prior-authorization layer specifically — not all AI use in health insurance — and uses "meaningful human review" as the operative standard rather than a flat prohibition on AI use. As with H.5253, whether S.443 advances in this General Assembly session is an open question.
Sponsored by Sen. Leber, S.963 was introduced February 26, 2026 and referred to a Senate committee. The bill adds a new Chapter 31 to Title 37 of the South Carolina Code, creating the Consumer Protections in Interactions with Artificial Intelligence Systems Act. The bill's central provision prohibits "algorithmic discrimination" arising from the use of "high-risk" AI systems, with definitions and safeguards that broadly track the framework Colorado adopted in its 2024 AI Act — the most comprehensive state-level consumer-protection AI law currently on the books anywhere in the United States.
S.963 is the broadest in scope of any AI bill currently in the South Carolina General Assembly. Where H.5253 targets education and S.443 targets health insurance, S.963 targets the cross-sector category of consumer-facing AI deployments that produce consequential decisions about housing, employment, credit, and other domains where algorithmic discrimination has been documented at the federal-policy level. If enacted, South Carolina would be among the first states beyond Colorado with a comprehensive consumer-protection AI statute. The bill's specific definitions of "high-risk" and "algorithmic discrimination," and the implementation timelines and enforcement mechanisms, are the substantive details to watch as the bill moves through committee.
Sponsored by Reps. Brandon Guffey (also a sponsor of H.5253), Sessions, Martin, and Brewer, H.5138 was introduced February 10, 2026 and referred to a House committee. The bill adds Chapter 80 to Title 39 (Trade and Commerce), establishing requirements and civil-action mechanisms specific to chatbot deployments.
Substantive provisions include definitions for chatbot, chatbot provider, chat log, input data, model, dark pattern, deidentified data, personal data, advertisement, and affirmative consent; an affirmative-consent requirement for certain chatbot practices, with detailed standards including clear stand-alone disclosure, plain-language drafting, accessibility for users with disabilities, and availability in commonly-used languages; and a private right of action for violations. The bill is targeted at chatbot deployments broadly — including but not limited to AI chatbots — and is part of a national wave of chatbot-transparency legislation introduced across multiple states in 2025–2026.
Sponsored by Sens. Kimbrell, Garrett, Rice, and Adams, S.1037 was introduced March 19, 2026 and referred to a Senate committee. The bill adds Chapter 81 to Title 39, focused specifically on chatbot use by minors. It defines covered entities as those with 500,000 or more monthly active users worldwide across all chatbots they offer, with explicit exemptions for internal tools, non-public tools, and customer-service-only deployments not offered to the general public.
The bill's core provisions require age verification and a limited-access mode for minors, define covered harms and covered incidents, and establish standards for parental notification and incident reporting. S.1037 is one of the more aggressive minor-protection AI bills introduced in any state in 2025–2026 and is structurally comparable to California's SB 243. The threshold (500,000 worldwide monthly active users) is calibrated to capture major commercial AI chatbots without imposing requirements on smaller deployments.
Two resolutions complete the General Assembly's AI-related activity in this session: H.5085, sponsored by Rep. McGinnis and a substantial number of co-sponsors, designates March 30 – April 2, 2026 as "AI Week" in South Carolina and encourages residents to learn more about AI; and S.225, a Senate resolution introduced January 15, 2025, advocates for open-source decentralized AI models as transparent alternatives to closed-source models and encourages policies that support open-source AI innovation while addressing privacy, safety, and accountability. Resolutions are non-binding but are useful indicators of where legislative attention is forming.
Five substantive bills, two resolutions, and a House committee with explicit AI legislative authority adds up to a state legislature that is — in volume and structure — taking AI legislation more seriously than its national reputation would suggest. The bills cluster usefully: two on a single sector (education and healthcare), one on cross-sector consumer protection, two on chatbot regulation. The resolutions add a literacy push and an open-source AI policy stance. None of the substantive bills proposes a flat prohibition on AI use; all of them require human oversight, disclosure, or consent as structural principles. Whether any specific bill advances out of committee in this General Assembly session is, at the time of writing, an open question. The directional message — that South Carolina is moving on AI legislation, not waiting — is clear.
Chief Justice John W. Kittredge issued an interim policy in 2025 governing the use of generative AI within the South Carolina judicial branch. ABC News 4 Charleston reported on the policy at the time of release. The National Center for State Courts published an analysis (AI Readiness for the State Courts 2025) that places SC's policy in national context.
The SC interim policy is recognizable as a working framework rather than a permanent rule. It addresses the use of generative AI by judicial-branch staff and addresses the standards expected of attorneys and self-represented litigants who use generative AI in filings. It is calibrated to the operational reality that generative AI tools are already in use across the judicial system and need governance now, not later. The policy is expected to evolve as the courts gain operational experience.
State-level AI policy in the U.S. exists in a federal context that has shifted significantly over the past two years. Several federal touchpoints affect SC policy:
Across the southeastern U.S. and the U.S. more broadly, state-level AI policy posture varies substantially. Some observations:
The combination of a published state strategy, a Center of Excellence, a designated legislative committee, and a judicial-branch interim policy puts SC ahead of most peer states. Texas, Virginia, North Carolina, and Georgia have varying subsets of these. Few states have all four.
H.5253's AI-in-education provisions are comparable to bills introduced in 2025–2026 in California, New York, Florida, Tennessee, and others. S.443's clinical-AI provisions are comparable to California's SB 1120 and bills in several other states. SC is neither leading nor trailing on sector-specific legislation.
States including California, New York, and Vermont have begun building or proposing public registries of state-agency AI deployments. SC's AI Strategy implies such a registry is coming, but as of this edition no public-facing registry yet exists. This is one of the more obvious gaps and is referenced in Chapter 8.
State-level AI policy is not solely the work of the three branches of state government. Several other actors matter.
The full picture of South Carolina's AI policy environment is not contained in the General Assembly or in the Department of Administration. It is distributed across all of the above. SCAIO's policy tracker will progressively expand to surface activity across these venues as it emerges.
The pieces are set. An AI Strategy. A Center of Excellence. A House committee with explicit AI authority. A judicial-branch interim policy. At least five AI bills in committee, plus two AI-related resolutions. A growing roster of state-agency activity.
What is harder to see — for citizens, for journalists, for legislators trying to think about AI broadly rather than one bill at a time — is the integrated picture. Each policy actor knows its piece; few have a complete view of the others'. That observation is the basis for several of the recommendations in Chapter 8.