Study: 75% of Dapp Transactions Are Now Made By Bots

The report - the largest-scale study of malicious bots in the EOS ecosystem - also found 51 percent of unique accounts and 75 percent of total transactions were driven by non-human accounts.

Using artificial intelligence, AnChain was able to root out repetitive or hyperactive accounts to determine that they were malicious bots.

In particular, during the study AnChain identified five Ethereum addresses behind an extremely sophisticated attack that employed 50,000 self-destructible malicious bots to steal $4 million over two weeks, by exploiting a contract flaw in a popular gambling game.

The study goes on to suggest that bot activity is a feature, not a bug, of decentralized blockchains.

Pseudonymous transactions "Leave the door open to bots going undetected for extended periods of time," in comparison to IP based internet accounts that are governed by a central authority, like ICANN or the SEC. "The decentralized nature makes blockchains even harder to defend than cloud systems," said Fang.

The study found that the most active Dapp, EOS, representing $480 million in weekly transaction volume, has only a small percentage of bot activity.

The second most popular Dapp exhibited the most bots, at around 1,900 out of the platforms 4,500 unique users.

The authors suggest, "This dynamic hints at the competitive nature of the Dapp world where the runner-ups are leveraging bots in order to augment overall ecosystem usage metrics."

The company suggests, in order to preserve legitimate competition, and draw earnest adoption, developers should institute automated quality assurance tests on their platforms, and discourage cheating through the implementation of malicious bots.

What's more, on slower Dapps, good bots could be programmed to interact with human players, who may not always find other gamers to play with "a bot player will be deployed to fill the void."