Assistive Chessboard; The Struggle over Artificial Intelligence in Healthcare
Chittari, Srikar, School of Engineering and Applied Science, University of Virginia
Powell, Harry, EN-Elec & Comp Engr Dept, University of Virginia
Norton, Peter, EN-Engineering and Society, University of Virginia
Developers strive to produce artificial intelligence (AI) for a range of applications from games to healthcare, but AI often encodes human biases. AI can defeat human experts in chess and help players of all levels improve their skills. Some physical, interactive chess boards train players by recommending moves, but most cannot reliably detect chess pieces. An assistive chess board was designed in which magnets improve piece detection. LEDs display the AI engine’s recommendations. A prototype was built and verified. It offers accurate, dynamic, and intuitive recommendations in real-time to one or more chess players. To resist discriminatory biases encoded in medical AI, U.S. advocates of healthcare equity use a variety of strategies. Most researchers propose improvements to the currently ineffective federal regulations on AI development. Others demand legal accountability in AI, voluntary ethics pledges, incentives, or AI audits.
BS (Bachelor of Science)
Artificial Intelligence, Healthcare, Chess, Stockfish AI, Algorithmic Bias, Human-Computer Interaction
School of Engineering and Applied Science
Bachelor of Science in Computer Engineering
Technical Advisor: Harry Powell
STS Advisor: Peter Norton
Technical Team Members: Ramie Katan, James Weeden, Iain Ramsey
English
All rights reserved (no additional license for public reuse)
2023/05/09
