tldr; For fifty years, pretyt much every major approach to measuring developer productivity has eventually been disavowed by its own creators. Lines of code, story points, velocity. Tom DeMarco wrote "you can't control what you can't measure" in 1982 and publicly retracted it in 2009. AI has made the problem worse and, weirdly, might also be the way out.
Some observations on a few ways different people actually gather feedback from humans in practice to improve LLMs. Sure I've missed some here, so let me know.
To expand on the analogy, what would a fully-self driving LLM look like? and would people even want that?
In most cases, I think the idea of AI that augments my own capabilities and helps me get things done more efficiently resonates more with me than AI that operates on its own on my behalf.
Good point on the terminology. What do you think the right terminology should be? LLMs is too much of a mouthful and is not as informative for the general public, imo. People are also using Foundation Models, which I rather like.
> Our collective awe-struck-ness has left us vulnerable to the fact that AI generation is, generally speaking, hollow and indirect.
This totally resonates with me. This is absolutely correct. Thinking about the future of work, there's much of what I do every day in my job that is hollow and indirect. And I would be totally okay if I could have something like ChatGPT do it for me.