Senate Bill 735 would create the North Carolina Artificial Intelligence Innovation Trust Fund within the Department of Commerce and appropriate $750,000 from the General Fund for fiscal year 2025-2026 to support its purposes. The fund could be used to provide grants or other financial assistance to companies developing or deploying AI models in key industry sectors, and to establish or promote AI entrepreneurship programs, including partnerships with research institutions and other support organizations.
The bill also establishes a broad AI governance framework for “covered models” and related derivatives, focused on preventing “critical harms” such as mass-casualty events, major cyberattacks on critical infrastructure, and other grave public-safety threats. It would require developers of covered models to adopt cybersecurity protections, written safety and security protocols, shutdown capabilities, annual reviews, third-party compliance investigations, compliance statements to the Attorney General, and rapid reporting of AI safety incidents. Operators of large computing clusters would also have customer-identification, recordkeeping, and shutdown obligations when their resources are sufficient to train covered models.
The bill would amend Chapter 143B of the General Statutes by adding a new Part on Artificial Intelligence Innovation, giving the Secretary of Commerce authority to administer the new trust fund and to develop regulations, guidance, and a tiered compliance framework for AI developers and computing-cluster operators. It would also authorize an advisory panel and require annual reporting to the General Assembly on AI workforce conditions, training gaps, and policy recommendations. Enforcement authority would rest with the Attorney General, who could seek civil penalties, injunctive relief, damages, and attorney’s fees; the bill also creates whistleblower protections for employees and limits private rights of action. In addition, the bill would impose new compliance, documentation, and reporting duties on AI developers and certain infrastructure operators, while expressly prohibiting funded projects involving mass surveillance, unlawful social scoring, discriminatory profiling, or deceptive content for fraud or election interference.
Based on the bill text, the overall tone is supportive of AI innovation but cautious about risk. The findings section emphasizes “responsible innovation,” stakeholder input, and an iterative approach to AI governance rather than immediate rigid regulation. Because there are no committee transcripts or recorded votes provided, there is no documented public debate or formal voting sentiment in the materials supplied. The structure of the bill suggests an attempt to balance economic development, research collaboration, and public-safety oversight.
The main points of potential contention are the breadth and specificity of the regulatory regime and the scope of state oversight over AI development. Developers and computing-cluster operators would face extensive recordkeeping, disclosure, shutdown, testing, and third-party audit requirements, which could be viewed as burdensome or difficult to implement, especially for smaller firms or open-source participants. The bill also gives the Secretary of Commerce significant rulemaking discretion to define covered models and update compute thresholds annually, and it authorizes the Attorney General to enforce the chapter with substantial penalties. On the other hand, the bill includes explicit limits on prohibited uses and whistleblower protections, reflecting concerns from civil liberties, consumer advocacy, and public-safety stakeholders about surveillance, discrimination, fraud, and catastrophic misuse.