Flutterby™! : LLM boosterism

Next unread comment / Catchup all unread comments User Account Info | Logout | XML/Pilot/etc versions | Long version (with comments) | Weblog archives | Site Map | | Browse Topics

LLM boosterism

2025-06-03 17:18:44.674949+02 by Dan Lyke 0 comments

Counter to my cynicism: Thomas Ptacek: My AI Skeptic Friends Are All Nuts. I remain skeptical, but a few things resonated, like:

If hallucination matters to you, your programming language has let you down.

Agents lint. They compile and run tests. If their LLM invents a new function signature, the agent sees the error. They feed it back to the LLM, which says “oh, right, I totally made that up” and then tries again.

This is one of the reasons I get frustrated with Python. On the other hand I've seen enough subtlety in C and C++ that I'm also not convinced there, and have my own complaints about Rust, but... good languages type. Hard.

Edit: Lobste.rs conversation

Edit 2: MeFi.

[ related topics: Interactive Drama Weblogs Invention and Design Software Engineering Monty Python Mathematics Artificial Intelligence Python ]

comments in ascending chronological order (reverse):