Since ChatGPT, the “assistant” frame has dominated how we think about LLMs. Under this frame, AI is a helpful person-like entity which helpfully completes tasks for humans.
In this post, I’ll argue for training AI to act as an “amplifier.” This new frame could help us sidestep some potential alignment problems, tethering the AI’s behavior more closely to human actions in the real world and reducing undefined “gaps” in the AI’s specification from which dangerous behaviors could emerge.
Read more here about different ways of thinking.
Basement Whatever 
Leave a comment