AI Frontier Institute

Creates a comprehensive statutory framework to address and regulate the use of artificial intelligence in the workplace, considering the interests of employers and employees.

View primary source →

Rhode Island H 7767 has not been enacted. The House Labor committee recommended on 2026-04-15 that it be held for further study. If enacted, it would add a new Chapter 5.2 to Title 28 of the General Laws, covering AI use and fair employment practices. Section 28-5.2-1 defines terms such as automated decision system (ADS), electronic monitoring tool, employee information, impact assessment, independent auditor, and meaningful human oversight. Section 28-5.2-2 would make it unlawful for employers to use electronic monitoring to collect employee information unless the tool is primarily used for a listed legitimate purpose. Those purposes are essential job functions, quality of goods and services, periodic performance assessment, legal compliance, health, safety and security, and administering wages and benefits. Monitoring would have to be narrowly tailored and least invasive, and it could not collect data when employees are off-duty. Employers would have to give prior written notice and get written acknowledgment, keep certain records for five years, and destroy data within 61 months unless the employee consents to retention. Section 28-5.2-2 would also bar: - facial recognition, gait, voice analysis, and emotion recognition tools; - monitoring in bathrooms and similar private areas; - monitoring of home workspaces and personal vehicles; - monitoring to obtain health or protected-class information; - monitoring to identify or punish activity protected under labor law; - adverse action based on continuous incremental time-tracking data, except for egregious misconduct; - adverse action based on undisclosed performance standards or monitoring without proper notice; - selling or disclosing monitoring data, with limited exceptions; - required implanted devices, personal-device apps, and location tracking beyond what is strictly necessary for essential job functions. Employers could not rely primarily on monitoring data for hiring, promotion, discipline, termination, or pay decisions. Those decisions would require meaningful human oversight and a human decision-maker who also considers other information. Employers would have to disclose to affected employees and their representatives, within 30 days, when such a decision was based on monitoring data. Employers using electronic monitoring, alone or with an ADS, would need an independent impact assessment. The assessment would be disclosed to workers, who could comment on, challenge, and bargain over the monitoring. Section 28-5.2-3 would protect employees from retaliation for refusing to follow AI or ADS output when specified conditions are met, such as holding independent judgment or a state-required license and having a good-faith belief that the output would cause harm. It would direct the Department of Labor and Training to issue implementing rules. It would also direct the Attorney General to issue rules on privacy and bias harms from AI and automated decision systems. Section 28-5.2-4 would prohibit penalizing employees who seek their rights or file complaints under the chapter, with penalties as provided in Title 28. Section 2 provides that the act takes effect upon passage.

← Back to the tracker