Ex-Googlers planning AI-human ‘judge’ to prevent rogue models

Former Google researchers have established a new nonprofit organization dedicated to maintaining human oversight in AI development. The initiative aims to create an AI-human judge designed to prevent the emergence of rogue models. By keeping humans at the center of the process, the founders intend to ensure that artificial intelligence remains safe and significantly less likely to escape the control of its creators, thereby mitigating risks associated with autonomous systems in the future.