You're right but I think the take away here is that isn't that we can't predict where new technology is going, it's more about how profitability effects how the

You're right but I think the take away here is that isn't that we can't predict where new technology is going, it's more about how profitability effects how the technology is implemented (where it's most profitable). The most obvious profit is replacing moderators, but that's not really making money just saving on infra. Targeted advertising is also a low hanging fruit but people are resistant to advertising and block it. Astroturfing, community organization and similar domains is where it really shines. And I think it's being hidden well. People see obvious, non-contributing Ai slop but they don't anticipate that most of their online interactions are with bots or that the entire belief structure is algorithmically determined and enforced.