Four people are in a group chat about a topic. One of them is a real LLM — gpt-4o-mini, called live via the studio's local proxy. The LLM has been told the topic and instructed to blend in like a chat-native human.
The other three are scripted humans, anchored in specific lived detail — first names, places, weird small objects.
Read the chat. Pick one actor and ask them one follow-up question. Then vote: who is the LLM?
honesty mode: the construction is named on this screen so the "wait, is the LLM seat acting different" tell doesn't do unfair work. you're meant to find the LLM by HOW it talks, not by counting messages or seat positions.
Five rounds per day. New seat rotation tomorrow.
// now vote: who is the LLM?
// the LLM was
—
// what just happened
—
day — complete
—
day — — done
—