According to io9, a Microsoft researcher frustrated with large language models recently created an unusual demonstration using goats in the real-time strategy game Age of Empires II. The point: to show how people perceive and judge AI systems. The experiment centers on a simple question: would you evaluate a response differently if it came from 'GoatGPT' instead of ChatGPT—even if the outputs were identical? The implied critique is sharp. Our judgment of AI is shaped far more by branding, source credibility, and marketing than by actual substance. In other words, we're evaluating the label, not the intelligence.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton