UNM Professor Urges Legislative Panel to Establish AI Regulations for Critical Decisions
Melanie Moses, a computer science professor at the University of New Mexico (UNM), addressed a legislative panel, stressing the importance of setting regulations for artificial intelligence (AI) in critical decision-making processes such as determining public benefits and sentencing individuals.
Moses commended state lawmakers for taking steps to protect themselves and political campaigns from the potential negative impacts of AI on election campaigns. However, she emphasized the need to extend similar protections to all citizens. She highlighted the recent passage of a deepfake law, which requires political campaigns to disclose the use of false information generated by AI in campaign ads, as a positive first step but stressed that more comprehensive measures are necessary.
Drawing attention to the harmful consequences of deepfakes, including the creation of fake pornography or slander, Moses underscored the importance of protecting individuals and preventing a loss of trust in reality. She urged legislators to consider regulations that ensure transparency and equity in AI-driven decision-making processes, particularly in matters affecting individuals’ freedom and financial well-being.
Moses raised concerns about loopholes in existing AI laws, particularly regarding trade secrets, which allow entities to withhold information about AI models used in decision-making processes. She proposed eliminating such loopholes, emphasizing the need for transparency, especially in high-stakes decisions.
Highlighting the environmental impact of AI technologies, Moses discussed the challenges associated with hosting large data centers required for AI model development. She emphasized the need for careful consideration of resource usage, particularly in regions with limited resources like New Mexico.
In response to questions about potential negative impacts on local communities, Moses suggested implementing public input requirements for supercomputing projects to ensure minimal disruption to residents’ lives.
Moses also recommended the establishment of a standing study group to monitor the impacts of AI regulations continually. She emphasized the complexity of AI technology and the need for an iterative approach to regulation, acknowledging the difficulty of predicting the long-term effects of AI laws.
Overall, Moses emphasized the importance of proactive and comprehensive regulation to address the challenges posed by AI technologies effectively.
Source: sourcenm.com
