HB 5222 clarifies and strengthens the Department of Consumer Protection's authority to investigate and enforce consumer protection laws. It specifically amends statutes to explicitly grant the Department and its board the power to issue subpoenas, administer oaths, compel testimony, and request documents during investigations. The bill also establishes immunity for staff acting in good faith and requires the state to cover legal costs for such actions. Additionally, it details enforcement mechanisms, including the ability to issue orders to stop violations and impose civil penalties up to $50,000 for violations of consumer protection statutes. The bill does not affect professional licensing fees or architecture regulations, which appear to be misplaced in the text.
This bill directs the Commissioner of Energy and Environmental Protection to evaluate ways to improve the state's permitting process by January 1, 2027. The commissioner must assess using artificial intelligence for preliminary reviews, explore automated tools for businesses and municipalities, and consider creating expedited review options with higher fees for certain permits. The report will also examine whether permit renewals without changes could be automated and have reduced fees. All findings and recommendations for new laws must be submitted to the General Assembly's relevant committees.
This bill requires the Department of Economic and Community Development to create a plan for an artificial intelligence small business program. The program aims to help small businesses adopt AI technology to improve productivity and product or service quality while fostering a competitive environment for AI development. The department must submit a report on the plan by January 1, 2027, including recommendations for any additional laws needed to implement it. The bill defines artificial intelligence as machine-based systems that make predictions or decisions affecting real or virtual environments.
This bill requires employers in Connecticut to inform job applicants and employees when automated systems are used to make employment decisions, such as hiring, firing, or performance evaluations. Companies deploying these systems must disclose what personal data is being collected, how long it will be kept, who will access it, and provide options for applicants to opt out of data processing. The law also mandates that employees receive written notice before any employment decision is made, explaining the purpose of the automated system and the nature of the decision being made. Developers of these systems must provide necessary information to employers to ensure compliance, unless they have a contract assuming those responsibilities. The protections apply to various automated processes including resume screening, online interview analysis, and predictive assessments used in workplace decisions.
HB 5312 establishes a private right for victims and a civil action for Connecticut's Attorney General to address unlawful sharing of synthetically created intimate images (like deepfakes). Victims can sue in Superior Court for damages, emotional distress, and attorney fees, while the Attorney General can pursue civil penalties up to $50,000 per day against online platforms that fail to remove such content after knowing it violates the law. The law specifically protects minors under 18, even if the image is synthetic, and applies to platforms like social media or apps that host user-generated content. It takes effect October 1, 2026, and complements existing legal remedies without limiting other rights.
SB 86 establishes data governance structures for Connecticut's executive branch agencies to improve data management and transparency. It requires each agency to appoint an "agency data officer" and creates a state "Chief Data Officer" to develop a biennial data plan, standardize data practices, and create an online open data repository. The bill mandates agencies to inventory high-value public data and publish open data (after privacy safeguards), with a specific provision (section j) directing the Chief Data Officer to identify data suitable for AI systems by 2028, while ensuring policies prevent discrimination. The law focuses on data infrastructure and access - not regulating AI technology or developers - and takes effect July 1, 2027.
SB 5 (AN ACT CONCERNING ONLINE SAFETY) requires subscription-based AI providers (e.g., companies offering AI tools via paid plans) to give consumers clear, written disclosures about subscription terms before signing or renewing. This includes detailing any usage limits, such as restrictions based on user behavior or changes to prior terms. The bill also establishes new safety rules for "frontier developers" of advanced AI systems ("foundation models"), defining "catastrophic risk" as scenarios where AI could cause mass harm (e.g., aiding weapon creation or severe physical injury) and mandating risk assessments by covered employees. It does not ban specific AI uses but sets transparency and safety protocols for high-risk systems. The law takes effect October 1, 2026.
HB 5342 bans the distribution of AI-generated or manipulated images, audio, or video (called "deceptive synthetic media") during the 90 days before an election if it misleads people into thinking a real person said or did something they didn’t. It specifically targets content intended to harm a candidate or influence election results without the depicted person’s consent. The bill allows exceptions for news organizations (like TV stations or newspapers) that add clear disclaimers (e.g., "This video has been manipulated") and cite the original source, provided they follow specific formatting rules. This law directly affects political campaigns, social media platforms, and media distributors during election seasons, aiming to prevent deceptive content from swaying voters.
