Describing my vision for AI as a "quiet plot" is funny, given that I have published a 60 page paper on it with numerous talks, posts, tweets... The "frightening" part is simply wrong, since the architecture I propose is a way to guarantee that AI systems be steerable and aligned.

1970s: Let's scare the heck out of people about nuclear energy, so that instead of zero-emission power plants we'll use lung-darkening, climate-warming coal and oil plants, killing millions in the process. 2020s: Let's scare the heck out of people about AI, so that instead of amplifying human intelligence and solving the world's biggest problems, we'll make sure progress slows down to a crawl.

LLMs have no physical intuition as they are trained exclusively on text. They may correctly answer physical intuition questions *if* they can retrieve answers to similar questions from their vast associative memory. But they may get the answer *completely* wrong 1/