> It seems like LLMs are great for the kind of generic work you'd usually pawn off to a new associate or paralegal

This seems to be the case with most industries that LLMs are impacting, coding, data input, legal etc.

Whilst it's great to get the time saving from not delegating to a junior (LLMs are much faster at the basic level 1 work) there's going to be this big knowledge gap that you really only gain from actually doing the mundane boring shit.

A lot of this _should_ be automated away, but we really need to think about how the next group of people are going to learn the skills and nuance to applying them. I don't really have an answer to that one other than still giving that work to junior staff, but they will then just utilise an LLM to help complete the job, and rightfully so, I'd have done exactly the same thing.

I think OpenAI tried doing this with some early chatgpt settings with "learning mode" or whatever that is. But an LLM just telling you what to do will never teach someone, there was something about having to go out and hunt for the answer through the sea of crap responses or wrong SO questions that really taught you something. How do you replicate that learning with an LLM and how do you apply that to different industries like Law?

> but we really need to think about how the next group of people are going to learn the skills and nuance to applying them

That's what school is for. For the last 150 years we've automated a bunch of things, and then we send people to school to learn how those automations work, so they can make better automations.

Maybe law school will have to be a year longer. Or maybe some forward thinking firms will have new associates check the work of the LLM to gain that experience.

This is kind of a new complaint. Places like adp have being trying to automate away mundane hr tasks. Oracle and SAP have been trying to automate away the mundane admin tasks. Siemens does it for OT. There's a huge long tail of trying to solve this as if it were a problem in almost every field. This has been going on for decades and no one complains that we will lose expertise in these fields if the drudge work goes away. But bring in an AI... and everyone is suddenly concerned about corporate drudgery.

Anyone who does not do the grunt work of learning how things work at a granular level will run into trouble when their agent suggests "let's re-write the firmware!" and the user says "sure, whatever that means", the model will get stuck or break something, and there will be no one around to understand what is going on.

With strict version history that may be okay, but hopefully their agent already thought of that for them too!

I don't think toiling away at mundane tasks is the only way to build expertise in a field, or even a particularly good one. Today's farmers barely touch hand tools anymore and they're able to produce crops just fine.

I've been having an identity crisis since generative AI started that is basically, "how am I going to learn anything with my ADHD brain when AI can do 10% of it immediately". That initial 10% would take me weeks or months to achieve on my own, and AI can do it right now, and that makes it even harder for me to bother. I hate it. I badly wish to do it myself because then I can do more than 10%, but I can't seem to push myself now that AI is right there.