Comment by thetwentyone
6 days ago
The author posits that people don't like using LLMs with Rust because LLMs aren't good with Rust. Then people would migrate towards languages that do will with LLMs. However, if that were true, then Julia would be more popular since LLMs do very well with it: https://www.stochasticlifestyle.com/chatgpt-performs-better-...
Does the linked study actually check that the LLM solves the task correctly, or just that the code runs and terminates without errors? I'm bad at reading, but the paper feels like it's saying the latter, which doesn't seem that useful.
I mean, just to steelman the argument, the "market" hasn't had time to react to what LLMs are good at, so your rebuttal falls flat. I think the original statement is more a prediction than a statement of current affairs.
Also, the author didn't say that "ease of use with LLMs" is the _only_ factor that matters. Julia could have other things wrong with it that prevent it from being adopted.