Technology ❯ Artificial Intelligence ❯ Natural Language Processing ❯ Language Models

ChatGPT

Studies Flag Ongoing Chatbot Errors as OpenAI Publishes Mental-Health Usage Data

Vendor-reported safety gains contrast with new findings of widespread hallucinations plus fabricated citations.

Karpathy Releases Nanochat, a $100 ChatGPT‑Style LLM Speedrun, as Community Publishes Hugging Face Build

Sam Altman Says AI Talk on X and Reddit Now ‘Feels Very Fake’

OpenAI Introduces Guided Learning Study Mode in ChatGPT

Humans Adopt ChatGPT’s Preferred Vocabulary in Spoken Language, Preprint Finds

ChatGPT's Voice Interface Misinterprets English Queries as Welsh