Fly Brain Successfully Made to Engage in Crypto Trading

cryptonews.ruPublished on 2026-09-12Last updated on 2026-09-12

Abstract

Researchers have conducted an experiment using a digital connectome (brain map) of a fruit fly, the Drosophila MaleCNS v1.0, containing 166,700 neurons and 25.6 million synaptic connections. In this project, real-time Bitcoin ($BTC) to US Dollar Coin ($USDC) price data from Coinbase was converted into an RGB visual signal. This signal was fed into the fly brain model, with "buy" actions encoded as "sweet" and "sell" as "bitter." A fixed neural reader then translated the brain's activity into trading commands—buy, sell, or hold—which were routed through an intermediary service to place spot orders on Coinbase Advanced. The experiment ran in a safe, simulated environment with virtual $100, not real money. The programmer noted that real-money trading is technically possible by creating a limited portfolio (max $100 USDC) with restricted API keys. However, the fly's model would be capped at trades of $10 each, with a maximum of 24 trades per day, no leverage, no shorting, and trading halted after a $20 portfolio loss. The system incorporates a learning mechanism: profitable outcomes stimulate 15 dopamine-like neurons (PAM11), while losses activate two aversive neurons (PPL101), encouraging the model to avoid unprofitable trades. Despite this, the developer acknowledges that the fly brain has not yet demonstrated an ability to trade profitably; synaptic changes do not equate to a successful trading strategy. This remains a research experiment. In related news, a Solana-based bo...

The experiment is based on the scientific brain map of the fruit fly Drosophila MaleCNS v1.0. The brain has 166,700 neurons and 25.6 million synaptic connections, explained the programmer. Public Coinbase prices for the $BTC-$USDC pairs are converted into an RGB graph, which is fed to 3,335 brightness inputs and 811 color inputs of the R8 neural network. Purchases are encoded as 'sweet,' and sales as 'bitter.' A fixed neural reader converts brain activity into a command: buy, sell, or hold. Then, a middleware service between the neural network outputs and the real trading system checks limits and places spot orders through Coinbase Advanced.

The project operates in safe mode, without real money. The prices for the $BTC-$USDC pair were real, but the fly traded with virtual $100. Trading with real money is also possible: for this, according to the programmer, one can create a separate portfolio with a balance of no more than 100 $USDC and generate an API key with viewing and trading rights, but without withdrawal permissions. The neuro-fly, the developer advises, is capable of executing trades worth a maximum of $10 including fees, with no more than 24 such trades per day. The model cannot use leverage, open short positions, and if the portfolio drops by $20, new orders are blocked.

Moreover, the fly learns trading gradually: with a positive portfolio result, 15 dopaminergic PAM11 neurons are stimulated; with a loss, two aversive PPL101 neurons are stimulated, causing the fly to develop an avoidance reaction to such trades. However, demonstrating that the fly's brain learns to trade profitably has not yet been possible, admitted Wormuth. Synaptic changes by themselves do not prove the ability for profitable trading—this is a research experiment, not a ready-made strategy.

Earlier, a bot on the Solana blockchain managed to turn 0.227 $USDC into approximately $696,000 in a few seconds using a single transaction. The profit was made thanks to a sharp drop in the price of the meme coin ANB.

end-content

Related Questions

QWhat is the scientific basis of the experiment that allowed a fruit fly's brain to perform crypto trading?

AThe experiment is based on the scientific brain map of the fruit fly Drosophila MaleCNS v1.0, which contains 166,700 neurons and 25.6 million synaptic connections.

QHow is the cryptocurrency price data transformed into input for the fly brain neural network?

APublic Coinbase prices for the BTC-USDC pair are translated into an RGB graph. This graph is then fed into 3,335 brightness inputs and 811 color inputs (R8) of the neural network.

QHow are trading actions (buy/sell) encoded for the fly's brain in this experiment?

AIn the experiment, purchases are encoded as 'sweet,' and sales are encoded as 'bitter' for the fly's brain.

QWhat mechanisms are used to teach the fly's brain to trade, and has it been proven profitable?

AThe fly learns through reinforcement: positive portfolio results stimulate 15 dopaminergic PAM11 neurons, while losses activate two aversive PPL101 neurons, creating an avoidance response. However, the programmer admitted it has not yet been demonstrated that the fly's brain can learn to trade profitably.

QWhat are the strict safety and operational limits imposed on the fly's trading system when using real money?

AIn a real-money scenario, the system would use a separate portfolio with a balance no greater than 100 USDC. Trades are capped at a maximum of $10 including fees, with no more than 24 such trades per day. The model cannot use leverage, open short positions, and new orders are blocked if the portfolio drops by $20.

Related Reads

Mercado Bitcoin begins post-hack reopening without a timeline for withdrawals

Mercado Bitcoin, one of Brazil's largest crypto exchanges, began restoring read-only access to customer accounts in phases on Monday, a week after attackers drained an estimated $49 million from the platform. Access started at 19:00 UTC, with all accounts expected to be viewable by 21:00 UTC, showing holdings as they stood on Nov. 2, the day of the hack. The exchange has yet to say when customers will be able to withdraw funds. Fabricio Tota, vice president of new business, said the exchange hoped read-only access would reassure customers that their assets were still recorded on the platform. He explained that the company opted for "a more conservative approach" to reactivating accounts "to ensure our return happens with the utmost security." The caution follows concerns from blockchain security analysts that attackers were able to access addresses, draining funds through standard transfers rather than a smart contract exploit. Blockchain analysts traced some of the stolen funds to customer deposit addresses, but Mercado Bitcoin says customers won't bear the losses. Tota said the vast majority of assets were never even touched by anyone and were not involved in the incident. On Nov. 4, Mercado Bitcoin pledged to cover all losses with company funds, bolstered by a $20 million investment from stablecoin issuer Tether in July. Tota highlighted that a portion of the stolen funds were also frozen in the hours after the company announced the attack. The next phase of recovery will require users to confirm their registration details before they can move funds. The exchange plans to restore withdrawals via Pix, Brazil's instant payment system, initially allowing customers to withdraw only to accounts in their own name. Tota gave no date for either step. Mercado Bitcoin has yet to explain how the attackers gained access to its wallets. Blockchain data company Bitquery found that as of Nov. 4, wallets linked to the attacker held most of the stolen funds that had escaped freezes, including 106.07 BTC.

cointelegraph2h ago

Mercado Bitcoin begins post-hack reopening without a timeline for withdrawals

cointelegraph2h ago

Avalanche’s AVAX leads crypto gains following NYSE tokenization deal

Avalanche's AVAX token rose as much as 10% on Monday, becoming the top gainer among the 100 largest cryptocurrencies, after the New York Stock Exchange announced plans to use Avalanche's blockchain to settle trades on a new platform for trading tokenized stocks around the clock. AVAX later eased to $8.86, still up over 8% on the day, with a market cap of around $3.95 billion and $439.7 million traded in the past 24 hours. Avalanche's total value locked stands at $633 million, up 4% over the past day. NYSE's head of digital issuance, Dennis Lee, said the exchange is examining how blockchain technology can expand its equities offerings and is picking the right technology for each job. The deal underscores Wall Street's growing effort to use blockchain to make stock trading faster and enable trading outside regular market hours; traditional US markets are open 9:30 am to 4:00 pm EST on weekdays, with trades often taking at least one business day to settle. The NYSE first announced plans for the platform in January, and Ava Labs revealed in September that the exchange had spent about a year testing Avalanche, though the final selection was not confirmed until Monday. Avalanche will record stock trades and help complete transactions on the platform, which will let investors buy and sell digital versions of stocks across different blockchains. Lee said the settlement ledger needs to be fast, reliable, permissioned, cost-predictable, and backed by a large developer community—all of which Avalanche meets. The platform could launch before the end of 2026 pending regulatory approval. Ava Labs chief business officer John Nahas said first conversations with the NYSE began nearly a year ago and called it an important step toward blockchain becoming foundational financial infrastructure. The news comes as other major firms invest in blockchain for faster, round-the-clock trading: Nasdaq plans to invest $100 million in Kraken's parent Payward to launch tokenized stocks by Q2 2027, and the DTCC launched a service in October allowing firms to turn traditional securities into digital tokens. The NYSE deal follows other efforts to bring traditional assets onto Avalanche, including Progmat moving over 452 billion yen in digital securities onto a dedicated Avalanche blockchain in July and Aave Labs' plans for a lending service using tokenized assets as collateral.

cointelegraph3h ago

Avalanche’s AVAX leads crypto gains following NYSE tokenization deal

cointelegraph3h ago

Trading

Spot
活动图片