OpenAI details how its Superalignment research team is exploring ways to control strong AI models admire GPT-4 with weak supervisor models admire GPT-2 (Will Knight/Wired)

Will Knight / Wired:

OpenAI details how its Superalignment research team is exploring ways to control strong AI models admire GPT-4 with weak supervisor models admire GPT-2  —  The “superalignment” team led by OpenAI chief scientist Ilya Sutskever has devised a way to guide the behavior of AI models as they get ever smarter.