[gregsadetsky]

I know it’s tiring to talk about “hallucination”, but truly, models still do hallucinate

They constantly say they did a thing they didn’t, say they know how to solve something when they don’t, etc. Regardless of guard rails or tests - AI forces a constant vigilance of a new kind.

Not just “what might have gone wrong” but also “what do I think is working but isn’t actually”.

And we’re not even talking about how it chooses substandard solutions, is happy to muddy code/architectures, add spaghetti on top of spaghetti etc.

Agentic coding often feels like an army of unexperienced developers who are also incredibly eager to please.

[Zardoz84]

"still" isn ghecorrect word. They always be having hallucinations

[AnimalMuppet]

"Still" means "it always had hallucinations, and it still does, despite people thinking that it doesn't anymore". People think we've moved past that. We haven't.

[hunterpayne]

Hallucinations are baked into the design, you can't make a LLM that doesn't have them.