I remain perplexed that everyone is so focused on using LLMs to automate software engineering, when there are language-based professions (like Spanish tutor, in your example) that seem more directly threatened by language models. The only explanation I've heard is that the industry is so excited about reducing spend on software engineering salaries that they're trying to fit a square peg into a round hole, and largely ignoring the square holes.
> The only explanation I've heard is that the industry is so excited about reducing spend on software engineering salaries that they're trying to fit a square peg into a round hole, and largely ignoring the square holes.
I think that's really just it, and I agree with you. There are many other areas LLMs can, and should, be more useful and effort put toward both assisting and automating.
Instead, the industry is focusing on creative arts and software development because human talent for that is both limited and expensive, with a third factor of humans generally being able to resist doing morally questionable things (e.g., what if hiring for weapons systems software becomes increasingly difficult due to a lack of willingness to work on those projects, likewise for increasingly invasive surveillance tech, etc.)
We're rushing into basically the opposite of what AI should do for us. Automation should work to free us up to focus more on the arts and sciences, not take it away from us.
I think it's because software engineers are the only group that can unanimously operate LLMs effectively and build them into larger systems. They'll automate their own jobs first and move on to building the toolkits to automate the others.
language-based professions like translation have been dying for years and no one has cared; they're not about to start now that the final nail's been put in the coffin.
