Comment by samrus
1 day ago
> Consumers would just use the underlying foundational models directly to solve their problem
Whether this could happen is a critical question.
What ive found currently is that its not possible because the non-determinism of the model means you cant trust it, and these tasks are such that people expect determinism. Even if they dont then they dont tolerate the kind of stupid mistakes these models make
What could happen is that the model becomes good enough to be near deterministic, or as deterministic as humans. I dont know if the current next token prediction foundation is sufficient for that. But who knows, maybe it is
If you want determinism, you ask Claude Code to write you a computer program to the desired specs, and then run that program over and over again, instead of renting the software from a SaaS vendor. I've done that a bunch - I'll show Claude the inputs, I'll tell it what output format I want, tell it the tech stack, and then have it write the program. Then I'll ask for a systemd service or crontab entry to run it repeatedly.
For a lot of tasks, classical deterministic computer programming is better than an LLM. It's just that LLMs are very good at classical deterministic computer programming now.