That's absolutely correct. I use LLMs heavily in my work and open source projects. These tools are effective, they can solve hard problems and they allow me to work on a wider range of tasks than could before.

But, they're also jagged in terms of functionality. When you work with a human, you can learn what their core competencies are, and then if you give them a task that falls within that domain, you can be reasonably sure they'll finish it correctly. That's not the case with LLMs. It might do one task brilliantly, and a next similar task, it just shits the bed on.

And since it has no understanding of the task in a human sense, it can't self correct, learn or improve. All its doing is stringing tokens together based on probability. So, you need a human in the loop to review what it's doing, to correct it when it makes mistakes, and to define the actual goals. My experience is that doing all that properly ends up taking up a lot of time, so your actual productivity gain per person aren't all that significant.

Companies that try to replace humans with LLMs will soon find that they end up with a whole bunch of code that doesn't actually work, and they have no hope of fixing. The double edge of LLMs is that they're really good at generating a lot of wrong code really fast.