FutureBit Tests a Fruit Fly Brain for Bitcoin Mining

Colored-pencil editorial illustration of a fruit fly neural connectome simulation alongside a conventional Bitcoin ASIC miner.
Colored-pencil editorial illustration of a fruit fly neural connectome simulation alongside a conventional Bitcoin ASIC miner.
Illustration: FutureBit’s HashFly experiment uses a digital model of fruit-fly neural pathways to explore Bitcoin-style hashing; it is a browser-based proof of concept, not a biological ASIC replacement.

FutureBit, the company behind the Apollo line of Bitcoin mining hardware, is testing an unusual question: can the wiring diagram of a fruit-fly brain be used to explore a radically different way of performing Bitcoin-style hashing?

The experiment is called HashFly. It is a browser-based proof of concept built around a digital reconstruction of neural pathways from the MaleCNS v1.0 fruit-fly dataset. Current demonstrations simulate 2,914 neural traces and connect their activity to simplified double-SHA-256 work. Recent reporting from Tom’s Hardware and Decrypt confirms that the project runs on conventional computer hardware and is not competitive with a modern ASIC miner.

HashFly Is a Simulation, Not a Biological Miner

The distinction matters. HashFly does not contain a living fruit-fly brain that is mining Bitcoin. Instead, software simulates selected pathways from a reconstructed connectome—the map of how neurons are connected—and uses that activity as part of an experimental hashing process.

The current demonstration also operates at a vastly easier target than Bitcoin’s live network. Tom’s Hardware observed roughly 100 kH/s in the browser proof of concept. By comparison, FutureBit’s compact Apollo III ASIC is rated by the company at 18 TH/s. That gap is enormous and shows why HashFly should be viewed as computing research rather than a new commercial mining machine.

Where the 1 W/TH Claim Comes From

FutureBit says that if the concept could eventually be scaled using real organic neurons, it could theoretically reach about 1 watt per terahash. The company has described that as roughly ten times the efficiency of leading 3-nanometer silicon ASICs.

That number is not a measured HashFly hardware result. Reporting on the experiment says the estimate is based on the total power consumption of a fruit fly and an assumption that its neurons could all be dedicated to continuously performing hashing functions. There is currently no demonstrated organic Bitcoin miner operating at 1 W/TH.

Why the Experiment Still Matters

Bitcoin mining has become a specialized semiconductor problem. Modern miners use application-specific integrated circuits designed to perform SHA-256 calculations with as little electrical energy as possible. Every major generation competes on hashrate, joules per terahash, chip fabrication, power delivery and cooling.

HashFly asks a more fundamental question: does the computation have to remain silicon-based forever? Biological neurons operate very differently from digital logic gates. A useful organic computing platform would need to overcome major challenges involving reliability, speed, programmability, interfaces, manufacturing and repeatability before it could compete with an ASIC.

FutureBit says it intends to expand the simulation to more of the available neural dataset and publish additional findings. Those results will be more useful than the headline efficiency estimate because they can show how the architecture behaves as the simulated neural system becomes larger.

Bitcoin Mining Is Becoming a Computing Laboratory

The project illustrates why Bitcoin mining can be useful as an engineering benchmark. SHA-256 provides a clearly defined workload, while miners relentlessly measure throughput against electrical consumption. That makes unusual computing architectures easy to compare conceptually with specialized silicon—even when, as with HashFly today, the experimental system remains many orders of magnitude away from practical mining.

HashFly does not replace the ASIC. Its value today is as a research experiment that tests how far engineers can stretch the definition of a computing substrate. If nothing else, it provides another way to investigate the relationship between computation and energy—one of the central engineering problems behind both Bitcoin mining and modern AI.

BitcoinVersus.Tech Editor’s Note:
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. If you would like to support to help further secure the integrity of our research initiatives, please donate here: 3C9o19EH5HSiwEPyCTmEKzxhNCbo2X6TTb

https://x.com/1BitcoinVersus/status/1937006164555993338

BitcoinVersus.tech is not a financial advisor. This media platform reports on financial subjects purely for informational purposes.

Leave a comment