Everyone here is assuming that a very large LLM is their goal. 5 years ago, transformer models were not the biggest hype in AI. Since they apparently have a 10

Everyone here is assuming that a very large LLM is their goal. 5 years ago, transformer models were not the biggest hype in AI. Since they apparently have a 10 year plan, we can assume they are hoping to invent one or two of the "big steps" (on the order of invention of transformer models). "SSI" might look nothing like GPT\d.