Microsoft researcher builds a working neural network out of goats in Age of Empires II to critique AI science
A Microsoft researcher built a neural network out of goats. Here's why it matters for how you evaluate AI research.
Why it matters
A provocative critique of AI research methodology: a meta-analysis of 315 papers reveals that over half assume language models possess human-like traits before experimentation begins. The goat demonstration is a thought experiment exposing how interface design and presentation bias can obscure fundamental questions about what AI systems actually do versus what we anthropomorphize them as doing.
The key facts
5 to know315 papers analyzed
>50% of papers assume language models have human-like traits pre-experiment
Neural network constructed from goats, bridges, and ice ramps in Age of Empires II map editor
Critique centers on methodology bias and anthropomorphization in AI research
Microsoft researcher authorship
Go to the source
The Decoderthe-decoder.com
Publisher excerpt: A Microsoft researcher built a working neural network out of goats, bridges, and ice ramps in the Age of Empires II map editor. What looks like a joke is a pointed critique of AI research methods. His analysis of 315 papers found that more than half already assume language models have human-like…