Senate Bill 747, titled the AI Learning Agenda, would create an Office of Artificial Intelligence Policy within the Department of Commerce and establish an Artificial Intelligence Learning Laboratory Program. The bill is designed to help North Carolina study how artificial intelligence is being developed and used, identify where existing law may be too restrictive or too weak, and generate recommendations for future legislation and regulation. It frames AI governance as an iterative, stakeholder-informed process that aims to support innovation while protecting consumers, civil rights, and the public interest.
The new office would be responsible for inventorying current state AI regulation, consulting with businesses, academics, civil society, and other states, and convening an advisory panel. It would also administer a learning laboratory in which AI developers and other applicants could participate in a structured testing and research program. The office would set a learning agenda, publish public participant registries and summary reports, and submit annual reports to the General Assembly beginning July 1, 2026. The bill also directs state agencies to compile inventories of AI technologies they use or are considering, including information about vendors, functions, data sources, security practices, bias testing, fiscal effects, and whether AI informs or makes decisions.
The bill would affect Chapter 114 of the General Statutes by adding a new article governing AI policy, research, and regulatory experimentation. It would not itself broadly regulate private AI use through direct prohibitions, but instead create a framework for identifying regulatory barriers and gaps, and for temporarily modifying certain enforcement conditions through regulatory mitigation agreements. Participants in the learning laboratory could receive limited, temporary mitigation from some regulatory requirements if they meet eligibility criteria, agree to reporting and safeguards, and operate within defined limits. The bill also makes clear that participation does not equal state endorsement and does not create state liability for participant conduct.
Because no committee transcripts or votes were provided, the overall sentiment cannot be measured from recorded debate or roll call history. Based on the bill text, the measure appears generally pro-innovation and pro-study, while also emphasizing consumer protection, privacy, cybersecurity, and bias mitigation. The structure suggests an attempt to balance industry development with oversight rather than impose immediate heavy-handed regulation.
The main points of potential contention are likely to be the scope of the new office’s authority, the burden of reporting and inventory requirements on state agencies and participants, and the extent to which regulatory mitigation could weaken existing protections during testing. Supporters would likely emphasize the bill’s evidence-based approach, transparency, and flexibility for emerging technology, while critics may worry about insufficient safeguards, administrative complexity, or the possibility that temporary waivers could expose consumers to harm. The bill also leaves significant discretion to the Office of Artificial Intelligence Policy, which could draw scrutiny from both industry and consumer advocates.
The bill would amend Chapter 114 of the North Carolina General Statutes by creating a new Article 11, establishing the Office of Artificial Intelligence Policy in the Department of Commerce, and authorizing that office to adopt rules, collect state AI inventories, administer a learning laboratory, and issue annual reports and policy recommendations. It would require state agencies to inventory AI systems and related details, and it would create a framework for temporary regulatory mitigation agreements that can modify certain enforcement conditions for approved participants. The bill would therefore expand state administrative oversight of AI while also creating a formal pathway for testing and studying AI technologies under controlled conditions.
No committee discussion or vote history was provided, so there is no recorded legislative sentiment to summarize from debate or roll call. From the bill text itself, the measure appears to have a generally favorable, reform-oriented tone: it is intended to encourage AI development and experimentation while addressing consumer protection, civil rights, privacy, cybersecurity, and bias concerns. The overall framing is collaborative and exploratory rather than punitive.
The most likely areas of contention are the breadth of the Office of Artificial Intelligence Policy’s discretion, the reporting and compliance obligations imposed on state agencies and participants, and the use of regulatory mitigation agreements that temporarily relax some enforcement conditions. Supporters are likely to favor the bill’s learning-laboratory model, transparency requirements, and focus on evidence-based policymaking. Critics may question whether the bill sufficiently protects consumers and civil rights during testing, whether it creates too much administrative burden, and whether the state should be granting any form of mitigation before AI risks are more fully understood.