AI risks, RLHF, and inner alignment: GPT stands to give the business world a major boost. But with everyone racing either to develop products that incorporate GPT or use it to carry out critical tasks, what dangers could lie ahead in working with a tool that applies essentially unknowable means (inner alignments) to reach its goals? This week’s guest Jérémie Harris speaks with Jon Krohn about the essential need for anyone working with GPT to understand the impact of a system comprising inner alignments that cannot – and may never – be fully understood.Additional materials: www.superdatascience.com/668)Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast) for sponsorship information.
