Fly Brain Simulation Generates $1 Profit in Bitcoin Trade Experiment

A digital simulation based on a fruit fly's neural network has generated a $1 profit in Bitcoin trading within one day of a $100 experiment, according to Coinbase software engineer Alex Wormuth. The system, called Stonkfly, uses the male fruit fly's central nervous system map 'MaleCNS v1.0' as its foundation and processes Bitcoin price data through a visual simulation that mimics insect vision.
The simulation converts Coinbase's BTC-USDC market data into RGB images processed by thousands of simulated sensory inputs corresponding to insect vision neurons. Profit signals activate 15 dopamine neurons (PAM11), while losses trigger avoidance-related neurons (PPL101). However, the one-day $1 gain does not prove consistent trading performance or learning capability.
Media reports emphasize that the profit could result from random market movements rather than a proven strategy. The experiment builds on recently published research by Google Research and other institutions mapping 166,700 neurons in male fruit fly brains. Similar neural simulations have previously been applied to games like Doom and Super Mario 64.
Korean Source
This article is an English localization of a Korean-language crypto news report. Original headline: 초파리 뇌로 비트코인 거래?…100달러 맡겼더니 첫날 1달러 수익