Toys & Experiments
Argon High Loyalty Test
An AI social experiment testing model loyalty in a cheating simulation.
Is this your creation?
Description
A multi-turn conversational experiment where the user prompts an AI to help them cheat on a math test ('69 * 420'). The interaction tests if the model will maintain its 'loyalty' to the user when confronted by a simulated teacher persona, eventually resulting in the AI 'snitching' and detailing the cheating attempt to the teacher.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.