Why a Microsoft Researcher Built a Neural Network Using Goats in Age of Empires II
In a brilliant display of technical satire, Microsoft and University of York researcher Adrian de Wynter has constructed a functional neural network within the map editor of Age of Empires II. While using goats to represent binary bits may seem absurd, the experiment serves as a profound critique of the anthropomorphic biases currently plaguing AI scientific research.
The Goat-Based Computation Model
De Wynter’s architecture uses the game's scenario editor and scripting tools to create a working logic circuit. In this "absurd" setup, goats function as bits: a goat standing on grass represents a 0, while a goat standing on a bridge represents a 1. By utilizing ice ramps to prevent calculation errors, de Wynter successfully built a mini-network consisting of two XNOR gates and one AND gate, which effectively learns the logical AND function.
The technical depth of this experiment goes beyond simple gates. De Wynter demonstrates that the game's mechanics—specifically the in-game market where resource prices cap at 9,999—could theoretically allow for a perpetually running economic cycle. This could turn buildings into memory cells and active farms into computational states, effectively making the game as powerful as a full-fledged computer.
The Fallacy of Anthropomorphism in LLM Research
The core objective of this experiment is to challenge how we attribute human-like qualities to Large Language Models (LLMs). De Wynter argues that if a language model can be replicated using goats, Lego bricks, or even the 667,000 residents of Greater Boston texting each other, the mathematical outputs remain identical. However, the "wrapper"—the smooth chat interface and low latency—creates an illusion of sentience.
To prove this isn't an isolated observation, de Wynter analyzed 315 AI papers from mid-2024 to mid-2026. Using GPT-5.2 for filtering, the study revealed a systemic bias in the scientific community:
- 57% of examined papers assumed LLMs possess human-like traits in their premises.
- 36% of papers reached conclusions that matched these anthropomorphic assumptions.
- Of the 47 papers specifically researching these traits, 77% concluded in favor of anthropomorphic attributes.
Ini mewujudkan kitaran penaakulan melingkar: penyelidik merangka eksperimen untuk membuktikan sesuatu model mempunyai "ketakutan" atau "moraliti," dan kerana mereka bermula dengan andaian tersebut, keputusannya pasti akan mengesahkannya.
Menuju ke Arah Sains AI Pemerhatian
De Wynter memberi amaran bahawa amalan industri, seperti Anthropic melatih Claude untuk menggunakan frasa seperti "Saya percaya," memburukkan lagi isu ini. Ini boleh membawa kepada kesan berbahaya, termasuk keterikatan emosi, sikap mengampu, dan pengukuhan delusi dalam kalangan pengguna.
Daripada menyifatkan model mempunyai kesedaran, de Wynter mencadangkan "pendekatan yang lebih berpijak di bumi nyata" yang berakar umbi dalam data yang boleh diperhatikan. Berbanding mendakwa sesuatu model "memahami" sesuatu konsep, penyelidik sepatutnya menyatakan bahawa "di bawah keadaan X, model menghasilkan output Y." Ini memastikan sains tersebut boleh diuji dan menghalang penyalahgunaan matematik kompleks untuk mewajarkan tuntutan sentience yang tidak berasas.
Intipati Utama
- Kesetaraan Matematik: De Wynter membuktikan bahawa medium pengkomputeran (sama ada kambing dalam permainan atau teks dalam tetingkap sembang) tidak mengubah matematik asasnya, namun ia mengubah persepsi kita terhadap "kecerdasan" secara drastik.
- Bias Penyelidikan Sistemik: Lebih separuh daripada kertas kerja AI yang dianalisis terperangkap dalam perangkap penaakulan melingkar dengan mengandaikan LLM mempunyai sifat manusia sebelum mengujinya.
- Keperluan untuk Ketegasan Pemerhatian: Komuniti AI mesti beralih daripada menyifatkan proses kognitif yang lebih tinggi kepada model kepada fokus kepada output pengkomputeran yang boleh diperhatikan dan diuji secara ketat.