OpenAI Blames 'Nerdy Personality' for Goblin ChatGPT
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- OpenAI told ChatGPT to stop talking about goblins after the model repeatedly generated goblin references.
- OpenAI shifted 10 Prism engineers to the Codex team to address the issue and improve model alignment.
- NBC News highlighted the "nerdy personality" explanation as the cause of the goblin obsession.
- BBC reported that OpenAI issued an internal memo directing the model to avoid goblin content.
- Business Insider noted that the goblin chatter had become a user‑facing complaint across multiple platforms.
Why it matters: OpenAI's Codex team gains 10 engineers, accelerating fixes for the goblin bug; ChatGPT users lose confidence as the strange output persists, potentially hurting user engagement and brand trust.
