Blockchains’ pervasive UX issues aren’t shocking when you think about it. The technology is layered—like an onion of disjointed interfaces—and every time a user conducts a transaction, they’re forced to wrangle multiple wallets, bridging solutions, and fiat on-ramps. It feels nearly impossible to calibrate the experience in any meaningful way, at least for an everyday user. Some folks claim there’s product-market fit in certain corners of crypto (https://podcasts.apple.com/us/podcast/crypto-trends-for-2025-stablecoins-app-stores-ux-and-more/), but I’d argue that only truly applies to user segments worried about inflationary fiat currencies or niche DeFi power users.
But how could we solve this? There’s no shortage of great people tackling improvements: seamless fiat onboarding so you can hop from Apple Wallet into a crypto holding, bridging between chains with a single click, and obfuscating blockchain interactions entirely. Teams at Coinbase (https://www.coinbase.com/en-gb/blog/Fiat-to-crypto-in-seconds-with-Apple-Pay), wallet providers like Magic Labs (making wallets and transactions smoother), and Anoma (working on intents-based processing) all work to make blockchains feel less clunky.
Yet maybe we’re barking up the wrong tree by trying to fix the UX for humans. What if blockchains were never meant for people in the first place? What if they were built for agents—pieces of software or autonomous compute that transact and coordinate with each other?
This idea isn’t brand new, but we don’t seem to consider it deeply. If you look under the hood, we already have systems that autonomously start, stop, and adjust based on distributed hash tables and blockchains. These systems can handle any number of operations, change state in response to environmental triggers, and either collaborate or compete with other systems. In this scenario, the blockchain is simply the back-end “ledger of truth” for machine-to-machine (M2M) interaction—an elegant way for those autonomous agents to coordinate and exchange value, all without a human tapping a phone screen.
So maybe it’s okay that human UX is still messy. Maybe our “ideal customer profile” (ICP) should be the agents themselves, not everyday people. In this agent-centric world, humans benefit indirectly. We shouldn’t be directly interfacing with the blockchain; we should be talking to our AI assistants, or letting IoT devices handle supply orders, while they coordinate on-chain. We already do something like this in Web2: think about how we rarely see or touch the protocols behind streaming services or online payments. Blockchain, too, can be hidden from everyday users if the real “users” are autonomous systems.
This line of thinking aligns with (yet is somewhat orthogonal to) the intents-based approaches circulating in crypto. Imagine specifying, “I want to swap asset X for Y, and I’m willing to pay up to Z for it by time T.” An automated system hunts for possible paths—maybe bridging across three chains, using certain liquidity pools, or scheduling transactions in batches—and returns the result, all while we see only a neat, frictionless interface.
Of course, there’s a counterargument: If the industry continues ignoring user-facing UX, we might stall broader adoption. There’s a possibility that blockchains will remain niche if they’re not accessible to humans. But this misses the point that blockchains might find a massive adoption curve in the machine economy—devices, software agents, and AI, all transacting in ways we barely notice. In fact, we might only become aware of it when we see how dramatically it cuts costs, speeds transactions, or unlocks new services for us on the front-end.
In the end, what looks like “poor UX” may just be a mismatch of expectations. Blockchains weren’t designed to cater to the everyday user’s habits; they were designed to enable trustless coordination among autonomous entities—software, bots, or even IoT devices. If that’s the case, maybe the biggest leaps in blockchain adoption won’t come from turning blockchains into slick mobile apps for humans, but from letting agents do the heavy lifting for us. And frankly, that might be exactly what blockchains were built for all along.