OpenAI tells ChatGPT models to stop talking about goblins

The AI firm said that unlike previous model bugs, the issue "crept in subtly".

OpenAI's GPT-5.5 upgrade to ChatGPT and Codex appears to be going smoothly, especially compared to the rockier GPT-5.0 release last August. It turns out OpenAI got ahead of one issue that was brewing before the release of GPT-5.5 models: a goblin fixation. GPT-5.5 is specifically instructed not to obsess over goblins, gremlins, and other mythical creatures.

The system prompt for OpenAI’s Codex CLI contains a perplexing and repeated warning for the most recent GPT model to 'never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or other animals or creatures unless it is absolutely and unambiguously relevant to the user’s query.' The explicit operational warning was made public last week as part of the latest open source code for Codex CLI that OpenAI posted on GitHub.

'Goblins, gremlins, and other creatures... When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works. Anyone with even passing experience of using the latest LLMs knows to expect the unexpected. They can spit out some really random and often disturbing stuff. But ChatGPT's 'multiplying' goblin infestation is a bit more pathological than that.

PCWorld reports that OpenAI’s GPT models, including GPT-5.5, developed an unusual obsession with mentioning goblins and similar creatures in responses. This quirky behavior stemmed from a 'Nerdy' personality instruction encouraging playful language use, which became reinforced through AI training processes. The goblin references became so prevalent that OpenAI implemented a direct ban in its Codex app, illustrating the unpredictable nature of large language model training.

Goblins have entered (and now left) the chat. When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works. Everyone is talking about how powerful AI has become. But it is also known to make mistakes. Sometimes the 'glitches' are massive, such as Claude deleting a startup's entire database in 9 seconds; other times, the problems with AI are simply annoying. Take the current 'goblin glitch,' for example. Over the past few weeks, the internet has been fixated on the way ChatGPT started slipping the word 'goblin' into completely normal responses. Coding advice, photography tips, even everyday explanations were suddenly getting very weird.