Today at a social event I was talking to a woman who happens to be a client about what her son plans to study in college. He had switched his major from computer science to engineering on the widely-shared assumption that AI will continue to eat coding jobs.
The other assumption is that jobs like those of contractors (plumbing, HVAC, electricians, carpenters) will be better bets. Not sure abut that. As contractor firms get rolled up by PE firms I think a lot of six sigma and CMM will likely be applied to them to improve processes. Total headcount may drop as service levels rise and design and preventative care improves. Or maybe not. Honestly I'd be happy to be proved wrong and see small-time contractors continue to flourish.
As I think I have said before, if the big problem of AI is making sure AIs act in alignment with human goals and humans are oh so very far from being aligned, it stands to reason that the most important jobs of the future will be in leadership, diplomacy, management, listening, boundary-crossing... to bring ourselves into alignment with one another. How this will look economically, how people will get paid for doing these things, that I don't know.
As a codicil to this thought, I was listening to Dwarkesh Patel interview a sinologist on his podcast while I was driving in Chicago last week. The guy pointed out that pre-COVID we had about 300,000 Chinese students in the US in any given year but that these days the US is lucky to have 1,000 in China. This is a huge gap. Outward-facing, low level person-to-person contact is hugely important to mutual understanding at any level. It's not so much median Trump voters we need to figure out how to align with as people abroad, first and foremost in China right about now. But everywhere, really. It's a big fucking job. Is AI gonna do that for us?
No comments:
Post a Comment