Laws and Ethics of Using AI in Utility Operations and Management

This presentation will present a brand new summary of a project documenting laws and ethics of AI in the water sector and feedback from all 50 states on impacts to certified operators and uses in utility management. Artificial Intelligence (AI) has the potential to revolutionize the operation and management of public water systems, improving efficiency, reliability, and sustainability. However, the adoption of AI in this context raises important considerations regarding legal compliance and ethical implications. This presentation will explore the laws and ethical frameworks that govern the use of AI in public water systems, focusing on the operational aspects and decision-making processes. This presentation is developed through two lenses: (1) Impacts on Operator Certification presenting new research documenting the response from each state regulatory agency on use of AI within the job roles of certified operators, and (2) Impacts on Utility Management laws and governance breaking down use of AI by specific applications and associated risks. For example, the risk and governance of AI to predict asset risk (e.g. which pipe will fail next), is very different than the risk and governance of AI to automate treatment operations (e.g. chemical dosing optimization). Both of these are current applications of AI being used in the water sector. Laws surrounding use of AI within the US are evolving daily. A summary of all current laws governing use of AI within the 50 states will be presented. At the time of writing this abstract (Sept. 2023), there are currently 23 states with active or pending laws governing use of AI. No governance of AI use exists nationwide at this time. However, the Artificial Intelligence Act is significant to watch as it details four categories of AI Risk: (1) Unacceptable Risk AI systems pose a clear threat to the safety, rights, or well0being of individuals, (2) High Risk Significant level of risk, but not unacceptable (e.g. public services) where errors could have significant impacts on individuals or society, (3)Limited Risk lower risk in applications such as non-critical domains where potential harm to individuals or society is relatively low, and (4) Minimal Risk Negligible risk such as general consumer applications or entertainment. Ethical implications related to AI usage will also be discussed, emphasizing the need for transparency, fairness, and accountability. The presentation will address questions surrounding algorithmic bias, decision-making accountability, and the potential impact on disadvantaged communities. Additionally, it will explore ethical frameworks and guidelines that can be applied to ensure responsible and equitable AI deployment. The presentation will conclude by providing recommendations and best practices for utility managers involved in the implementation and oversight of AI systems in public water systems. By understanding the legal and ethical dimensions, participants will be better equipped to navigate the complex landscape of AI in water, fostering responsible and sustainable practices for the benefit of communities and the environment.