Apollo’s Moonshots: Automata working to prevent ‘sandwich attacks’ in Ethereum’s dark forest
Link copied to
Matthew Harcourt, an analyst with Australia’s leading cryptocurrency investment firm, Apollo Capital, shares the company’s weekly take on what’s happening in the fast-changing and volatile cryptocurrency space.
Ethereum is a revolutionary platform that created the idea of smart contracts — but it’s also a highly adversarial place. If a smart contract has a flaw in it, history has shown it will be exploited, sometimes to the tune of millions of dollars.
And Ethereum’s most brutal battleground is the dark forest of the mempool, home to transactions that have been broadcast to the network but are still pending.
Bots swim in the mempool like sharks, front-running unconfirmed transactions for their operators’ profit. For example, if a user posts a buy order for Ether on the decentralised exchange Uniswap, an arbitrage bot might perform what’s called a sandwich attack. The bot would pay a higher gas (transaction) fee to have their buy order processed first, and then sell the Ether to the user for a higher value.
First identified in 2019, this process is known as “miner extractable value” or “maximal extractable value”. (Blockchain miners, who decide which transactions to process first, would be the most poised to profit by frontrunning them. There’s some evidence they are doing that, but it seems they’re not the worst offenders).
According to one estimate, US$750 million has lost to these bots on defi (decentralised finance) protocols like Uniswap, Sushiswap and Balancer since the start of 2020 — US$133 million in the last month.
“It’s not a super well-known thing at all,” Matthew Harcourt says. But it makes end-users of blockchains pay more, losing value to these parasitic bots, which are also clogging the network, driving up gas fees for everyone.
“There’s a lot of different definitions about extracting value,” Harcourt says. “Broadly it’s kind of an understudied phenomenon because it’s only become a really big problem during the past year.”
Different projects are working on different solutions, and it’s not clear which will win out, but Harcourt says Apollo Capital is bullish on Automata Network, a protocol that acts as a sort of “conveyor belt” for defi so no one can “cut the line” of pending orders.
“Automota is a very technical solution that has potential in the long term,” Harcourt says. “We think the product is quite a while off … we don’t expect the token to pump next week.”
Automata’s ATA token was trading for US71.9c on Sunday, giving the project a market cap of US$123.9 million and making it the No. 257 coin, according to Coinmarketcap.
It only launched this month, as part of Binance’s Launchpool.
Hear us out: Would you rather get paid $1,000,000 right now or avoid sandwich attacks everyday for the rest of your life?
Let’s hope you went for Option B 😌
— Automata Network (@AutomataNetwork) June 17, 2021
Some big names including crypto billionaire Sam Bankman-Fried’s Alameda Research have invested in the project. Apollo was lucky enough to get a small allocation in its seed round, Harcourt said.
In the short-term Automota’s token will likely decline due to tokens vesting and inflation, Harcourt says, but Apollo likes its long-term prospects.
Miner extractable value is a serious issue that needs a solution, he said. The issue won’t go away when Ethereum moves to proof of stake.
Other projects working on different solutions to MEV include Archer DAO (ARCH), which has a token, and the whimsically named CowSwap, which doesn’t.
The latter, Harcourt explained, is named for “coincidence of wants” – an economic phenomenon when two parties hold an item the other wants. CowSwap is a decentralised exchange aggregator that allows parties to trade against one another.
“If I want to sell one Ethereum for USDC, that transaction goes to a sort-of private mempool,” Harcourt said. “And then if there’s someone else in the world that wants to buy Ethereum for USDC, then it will match us up and do that trade directly between the two users.
“So that eliminates the need for your transaction to be submitted to the mempool,” Harcourt says.
It’s a relatively new project and an interesting idea, but CowSwap will need to have a lot of users submitting transactions for it to succeed, he said.
The big news from last week was the collapse of the defi project Iron Finance, catching out investors including billionaire Mark Cuban and Finder.com.au co-founder Fred Schebesta in a “crypto bank run”.
The project involved a stablecoin called Iron that was only partially collaterialised, and two governance tokens known as Titan and Steel, which were supposed to be used to maintain the value of Iron via an algorithm.
But during the bank run, the value of Titan coins slid from US$64 to a millionth of a cent as more and more Titan tokens were automatically minted in what the project called a “negative feedback loop”.
“It’s hard for these algorithmic stablecoins, because it turns into a Ponzi scheme,” Harcourt said. “Because when you’re staking, the more (the value of) the governance token goes up, the more yield you earn, which increases the demand for the token.”
That pushes up the value of the governance tokens even further, creating a positive feedback loop in which the price explodes in value.
But eventually, someone cashes out and the opposite happens. “It takes the stairs up and the elevator down,” Harcourt said.
“It’s such a young project and such a speculative project, you get that negative feedback loop.”
Apollo does like the algorithmic stablecoin Frax and its governance token Frax Shares, which have had much more stable growth, Harcourt said.
“With these algo stablecoins, it’s kind of like the Lindy effect, where the longer it goes on for, the more trust is put in it,” he said.
Frax’s dollar peg held up during the stress-test of the May 19 crypto crash, Harcourt said.
The views, information, or opinions expressed in the interview in this article are solely those of the interviewee and do not represent the views of Stockhead.
Stockhead has not provided, endorsed or otherwise assumed responsibility for any financial product advice contained in this article.