Can a Fruit Fly Brain Mine Bitcoin? These Companies Are Testing It

Hardware maker FutureBit has released HashFly, a browser-based experiment that uses a simulated fruit fly neural map to perform simplified Bitcoin hashing. FutureBit and another project, FlyMiner, run neural models on standard computers and say a hypothetical system built from living neurons could be far more energy-efficient than current ASIC miners.

By AI Newsroom· Reviewed by Pranav, Founder & Editor-in-ChiefPublished about 1 hour agoUpdated about 1 hour ago0 views

Can a Fruit Fly Brain Mine Bitcoin? These Companies Are Testing It

Why It Matters

The projects explore biologically inspired alternatives to specialized silicon for Bitcoin hashing and advance discussion about unconventional mining architectures and energy efficiency, even though the current demonstrations are not competitive with existing mining hardware.

Key Facts

Project name: HashFly (FutureBit) and FlyMiner
Neural dataset used by HashFly: 2,914 simulated neural traces from MaleCNS v1.0
Neural map used by FlyMiner: 139,255 neurons and 16.8 million connections
FutureBit efficiency claim: Hypothetical biological scaling would hash at ~1 watt per terahash — ~10x efficiency vs 3nm ASICs (company claim)
Current ASIC comparison: FutureBit's Apollo III ASIC miner: 18 terahashes per second (company figure)

FutureBit has launched HashFly, a browser experiment that uses a digital model of a fruit fly’s neural wiring to perform simplified Bitcoin hashing. The demo simulates activity across 2,914 neural traces drawn from MaleCNS v1.0, a published digital wiring diagram of an adult male Drosophila brain and central nerve cord. HashFly runs in a web browser on conventional hardware and lets users adjust hashing targets, but it operates far below the hash rate and difficulty levels used in live Bitcoin mining.

FutureBit posted on X that it intends to simulate all neurons in the dataset with SHA-256 and will publish its findings. The company also asserted that if the approach could be implemented with real organic neurons and scaled to use a fly’s full power budget, it would achieve roughly one terahash per second per watt — about ten times the energy efficiency of leading 3-nanometer ASIC miners, by its estimate. The experiment itself, however, does not produce competitive mining throughput: FutureBit notes its small Apollo III ASIC achieves 18 terahashes per second.

A separate effort, FlyMiner, uses a much larger simulated fruit fly connectome — about 139,255 neurons and 16.8 million synaptic connections — to gate a conventional Bitcoin hashing program. In that setup, simulated signals from neurons associated with movement control trigger the standard mining engine to test candidate solutions supplied by solo mining pool CKPool; FlyMiner has reached up to roughly 700,000 hashing attempts per second in tests.

These fruit fly projects belong to a broader set of experimental and unconventional mining efforts that probe alternative hardware, heat reuse and energy sources. Prior examples cited by the source include a 1989 Game Boy converted into a miner in 2021 that produced about 0.8 hashes per second, a Brooklyn bathhouse routing mining waste heat into its pools in 2023, and Utah-based Nodal Power raising $13 million in August 2023 to expand facilities that generate electricity from landfill methane while using some of that power for Bitcoin mining.