Everything can always be more complicated, of course. For example: 1. The AI will never know the user well enough to predict what will be best for them. It w

Everything can always be more complicated, of course. For example: 1. The AI will never know the user well enough to predict what will be best for them. It will resort to treating everybody like children. In fact, many of the crude ways LLMs currently steer and censor are already infantilizing. 2. The users' benefit vs. "for your own good" as defined by a product vendor's financial interest is a scam that vendors have perpetrated for ages. Even the unsubtle version of it has a bunch of stooges and chumps that defend it. Things will not improve in the users' favor when it's harder to notice, and easier to pretend they're not malicious. 3. A bunch of Californians using the next wave of tech to spread cultural imperialism is better than China doing it, I guess. But why are those my options?