top | item 43735752

New OpenAI models hallucinate more

5 points| pixiemaster | 10 months ago |slashdot.org

2 comments

order

8474_s|10 months ago

Looks like latest chatGPT likes to hallucinate code segments that don't work or wrong API calls. Simpler instructions work fine, but at some point it starts to diverge and experiment.

jqpabc123|10 months ago

Is it the models or the fact that earlier hallucinations are starting to spread, accumulate and contaminate training data?

Performance degradation over time has been a recognized concern for a while now.