Microscope AI as a term refers to the proposal detailed here, though I agree that I don’t really understand the usage in this post and I suspect the authors probably did mean OpenAI Microscope.
We meant the linked proposal. Although I don’t think we need to do more than verify a GPT’s safety, this approach could be used to understand AI enough to design a safe one ourselves, so long as enforcing modularity does not compromise capability.
Microscope AI as a term refers to the proposal detailed here, though I agree that I don’t really understand the usage in this post and I suspect the authors probably did mean OpenAI Microscope.
We meant the linked proposal. Although I don’t think we need to do more than verify a GPT’s safety, this approach could be used to understand AI enough to design a safe one ourselves, so long as enforcing modularity does not compromise capability.