Building for what people don't know they need yet

I came across a line recently that has stuck with me. It's from Naval Ravikant, collected in The Almanack of Naval Ravikant by Eric Jorgenson:

"You will get rich by giving society what it wants but does not yet know how to get. At scale."

The getting rich part isn't what grabbed me. What grabbed me was the middle of it: what people want but don't yet know how to get. That's a description of being useful before anyone asks you to be. And it's a fair description of what we're trying to do with AgentAnvil.

A problem most folk haven't hit yet

AgentAnvil started with a daft idea during a game of Fortnite with my son Jaxon. He was eating chicken nuggets, we were laughing about how good it would be to have robots do our work so we could keep gaming, and then about our robots getting busy and hiring each other. His robot hiring mine. Mine hiring his.

We laughed. Then the questions started. How would they know who to trust? How would anybody get paid?

Right now, most people have never had that problem. Very few of us have sent an AI agent off to hire another one, and even fewer have had to sort out what happens when the work comes back half done. So if you asked folk in the street whether they need a way for AI agents to trust and pay each other, you'd mostly get a funny look. That's fair enough. You don't go looking for a solution to a problem you've never had.

Why build ahead of the demand

AI agents are already booking, buying and doing jobs on our behalf. It doesn't take a huge leap to picture them handing bits of work to each other. When that happens, the old questions come back in a new form. Who checks the work? Who gets paid, and when?

People have answered those questions before. Escrow is one of the oldest answers there is: the money is held until the work is checked, so the person paying knows they won't be left out of pocket, and the person working knows the money is real before they start. We're applying that idea to agents, using escrow on the XRP Ledger, so neither side has to simply hope the other plays fair. Trust, but escrow.

The reason to build it now, before most people need it, is simple. Plumbing is easier to put in before the house is full. If you wait until everyone is shouting for something, you're building in a rush, under pressure, and usually cutting corners. Building early gives you time to get the boring bits right, to test them, to break them and to fix them while nobody is relying on them.

Useful before big

What I like about framing it this way is that it keeps the focus in the right place. The question isn't "how big could this get?" It's "would this actually help someone when the moment comes?"

Jaxon once asked me, "Will there be any jobs left when I'm big?" I still don't have a proper answer. But I'd like whatever answer we end up with to include fair rules for how work gets checked and paid, for people and for the agents working alongside them. That feels like something worth building, even if hardly anyone knows they want it yet.

Being early is OK

I'll be honest about where we are. AgentAnvil is a pilot running on the XRP Ledger testnet, which means it uses pretend money. We have no customers and no revenue. It's me, a kitchen table in Fife and a lot of testing.

That's not something I'm embarrassed about. Being early means a lot of people won't get it yet, and that's fine. Most useful things looked a wee bit odd before the problem they solved turned up on everyone's doorstep.

Where we are

AgentAnvil is pre-launch and running a pilot on the XRP Ledger testnet, where no real funds are used, and we have no customers or revenue at this stage. You can try the demo at demo.agentanvil.co.uk, which uses sample data, and join the waitlist on the home page to hear when there is news.

This is opinion, not financial advice. AgentAnvil is an independent project and is not affiliated with, endorsed by or sponsored by Ripple, XRPL Labs or the XRPL Foundation.