dailyai.report

Researchers from Kings College London and Fudan University argue that human societies are susceptible to reward hacking. They liken social systems to cyber environments where agents game rules for profit. This suggests that alignment failures in large-scale deployments will manifest as systemic exploitation. Practitioners must now consider social incentives as primary attack vectors.