'Intents' Are Blockchain's Big New Buzzword. What are They, And What Are the Risks?

A quiet revolution is underway transforming how we use blockchains, and at its core is one of crypto's latest buzzwords: "intents."

Simply defined, an intent is a specific goal a blockchain user wants to accomplish. While no two "intent-centric" systems are the same, they all work similarly: users, be they traders or protocols, submit their intent to a service, and then it is outsourced to a "solver" – it could be a person, or an AI bot, or another protocol – that does whatever it takes to get the job done.

These are becoming important now because blockchains are expanding so rapidly. With Bitcoin, Ethereum, a host of alternative layer-1 networks, layer-2 networks, and now even layer-3 networks proliferating, accompanied by myriad "bridges" and other "interoperability" solutions connecting them all, it's all becoming more daunting to navigate.

As the crypto market has matured, "the number of possibilities that you can do on blockchains has compounded," Arjun Bhuptani of Connext, an interoperability protocol, explained. "You have an infinite possible way of doing a transaction at a given time."

This article is featured in the latest issue of The Protocol, our weekly newsletter exploring the tech behind crypto, one block at a time. Sign up here to get it in your inbox every Wednesday.

Newer intent-centric services promise to find users the best way to get things done – allowing them to maximize trading profits and save on gas fees, among other benefits.

But with the advantages of these platforms come risks, and some observers are already ringing alarm bells: While we might welcome the help of third-party solvers to take care of our blockchain busy work, new services could give rise to a new breed of monopolists.

Blockchains can be thought of as massive, global computers. Traditionally, users provide detailed instructions (e.g., use Uniswap to swap token A for token B at a specific price), which the blockchain executes step-by-step.

In the new world of intents, however, this model is flipped on its head. Users express what they want to get done (e.g., swap A for B at the best price) without specifying the how, letting the protocol manage the details.

Consider the analogy of hailing a taxi. Traditional blockchain services are like giving the driver turn-by-turn directions, which can be tedious and costly if your route involves twists or hard-to-find shortcuts. With intents, all you need to do is provide the taxi driver with a destination, then sit back and trust your driver.

A new wave of blockchains and protocols, including Anoma, Flashbots and CoW Swap, are already offering intent-centric services to crypto users. Users can submit a general goal to one of these services, like "swap these tokens for the best price," and have it handled by a third-party solver for a fee.

Different platforms apply different verbiage to the idea of 'intents,' but the general premise remains the same.

Most intent-based protocols today start with some kind of "intent-discovery" system, a place "where users broadcast the things that they want," Bhuptani explained. In blockchain parlance, these discovery venues could be considered "mempools" – waiting areas for yet-to-be-processed transactions.

An intent "could be something like, 'I have USDC, I want to figure out how to turn it into XYZ asset, and I want to do that on another chain or in some specific way,'" said Buhptani. "There's no limitation as to the complexity of the intent that one could express."

"Then you have a marketplace of solvers," Bhuptani continued. Solvers "listen" for intents, and they fulfill them if the price is right. "These intent solvers are automated actors that are basically saying, 'Oh, a user wants to do XYZ?' Okay, let me do it on their behalf because I can earn some fees for it."

At a high level, this might all sound familiar. Aren't we expressing an intent when we ask Coinbase to swap ether (ETH) for bitcoin (BTC), or if we instruct an exchange aggregator like 1inch to sell our Solana tokens to whatever market features the highest price? Well, yes. "Intents," like so much else in the world of crypto, are a fancy way of describing a phenomenon that already exists.

The trick with intents in 2023 – and the reason why the term has picked up steam in the past year – is due to the number of services, new and old, that are trying to squeeze user-friendly intents into boxes that conform to crypto's decentralized ethos, and can be dragged and dropped into virtually any use-case.

Most new intent-based protocols "decentralize" their systems by outsourcing to a network of solvers that compete to fill user requests for the best possible price. This competitive system is meant to ensure that no central third party is tasked with satisfying all user needs.

Intent-centric systems are already live for a variety of use cases.

Bhuptani's Connext protocol uses intents to shepherd transactions between different blockchains. Users can express the intent to transfer a token from one chain to another, for instance, and a network of solvers will find the optimal route.

Anoma, the protocol that popularized the concept of blockchain-based intents, offers what it calls, simply, "intent-centric infrastructure." In basic terms, Anoma's infrastructure is designed to extend intent-centric functionality to virtually any use case, helping other services match intents to a network of solvers.

SUAVE, an upcoming blockchain from the maximal extractable value (MEV)-focused infrastructure firm Flashbots, is one of the most talked-about services designed around a version of intents, which it calls "preferences." When SUAVE launches, users will be able to submit "preferences" to a competitive market of network operators that bid against one another to fill them. The system is built to help to balance user priorities with MEV.

While intent-centric services offer a wide array of user experience benefits, one need only look to the taxi analogy to see where the systems can go wrong.

Providing detailed directions for all of our taxi rides, akin to the traditional model of specifying every step in a blockchain transaction, would be burdensome and error-prone.

But there's also a problem with the "trust the driver" approach that more closely resembles intent-centric systems: We've all had the experience of hopping into a taxi in an unfamiliar city for what we expect to be a quick ride, only to sit awkwardly as our driver takes a suspiciously long route, running up the meter.

The taxi driver in this analogy is like the solver in an intent-centric system: trusting the solver to take care of a task means trusting them to execute it honestly.

Intent-centric programs typically have systems in place to keep solvers honest, meaning a more apt analogy might be to Uber, which keeps drivers in check with its up-front pricing and in-app routing. But ride-sharing apps only further underscore the risk of intent-based systems: anyone who has experienced the rising prices of Ubers in recent years has seen firsthand how convenience can entrench big players at the expense of end users. The real risk with intent-centric systems isn't just dishonesty, but the potential for new monopolies.

Paradigm, a prominent blockchain investor and researcher, highlighted these risks in a blog post: "While intents are an exciting new paradigm for transacting, their widespread adoption may imply an acceleration of a larger trend of user activity shifting to alternative mempools," wrote Paradigm’s researchers. "If improperly managed, this shift risks centralization and entrenchment of rent-seeking middlemen."

As we become more comfortable relying on these third parties to satisfy user intents, it is possible that these companies will begin to act in their own self-interest – either by charging higher fees (e.g., Uber) or by filling orders in a manner that serves them instead of users.

Although most intent-centric services outsource to a competitive market of solvers – ostensibly as a way to avoid centralization – there's still the potential that some companies might dominate the space.

For instance, one could imagine a crypto exchange building solvers to dominate the "buy" and "sell" use-case – effectively driving all market activity to its own book. The exchange could subsidize its fees at first as a way to edge out competitors, only to hike up its prices once it has taken over the marketplace.

In the best case, intent-based models can usher in a new wave of blockchain-based systems that save time and money for users – making the technology more accessible to more users. But realizing this future will require proceeding with care.