3 ms·
For a little while. But we’re already at the point where a layperson can feed back errors into the LLM well enough for most apps for it to fix the problems and
by dyauspitr 4mo ago
For a little while. But we’re already at the point where a layperson can feed back errors into the LLM well enough for most apps for it to fix the problems and those are the errors it misses, it gets most of them on its own.
- bigstrat2003 4mo agoWe most certainly are not at that point.