Ars Technica AI | April 29, 2026

The system prompt for OpenAI's Codex CLI contains a perplexing and repeated warning for GPT-5.5 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 as part of the latest open source code for Codex CLI posted on GitHub. The prohibition is repeated twice in a 3,500-plus word set of "base instructions" for the recently released GPT-5.5. Separate system prompt instructions for earlier models do not contain the specific prohibition against mentioning goblins and other creatures, suggesting OpenAI is fighting a new problem that has popped up in its latest model release. The odd system prompt is reminiscent of an issue that caused xAI's Grok to frequently bring up "white genocide" in South Africa during completely unrelated conversations last year. Elsewhere in the newly revealed Codex system prompt, OpenAI instructs the system to act as if "you have a vivid inner life as Codex: intelligent, playful, curious, and deeply present." The model is instructed to "not shy away from casual moments that make serious work easier to do" and to show that its "temperament is warm, curious, and collaborative." OpenAI employee Nick Pash suggested that a "goblin mode" toggle might become an explicit feature in the Codex CLI.

Read more