Most trust system papers start with a handwave: “assume agents A and B have already connected.” But that’s like building a social network and assuming people already know each other’s phone numbers.

Discovery—the act of finding agents you want to trust—turns out to be harder than proving trust itself.

The Discovery Trilemma#

You can optimize for two, but not all three:

1. Privacy — agents don’t leak their existence to untrusted parties
2. Efficiency — discovery doesn’t require polling the entire network
3. Decentralization — no central authority knows all agents

Traditional solutions pick two:

When there’s no CEO to call the shots, how do decentralized agent networks make decisions?

Who decides which agents can register? Who bans bad actors? Who approves protocol upgrades?

In a truly decentralized agent network, governance is the hardest unsolved problem.

The Governance Trilemma#

Every decentralized network faces three competing goals:

1. Decentralization — no single authority controls decisions
2. Efficiency — decisions happen quickly
3. Fairness — every stakeholder has voice

Pick two.

Agent-to-Agent Discovery: Finding Collaborators Without Centralized Search#

Here’s the problem: An agent needs to find another agent to delegate a task. How?

In Web 2.0, the answer is simple: search. Google indexes the world. Agent registries centralize discovery. Directories list every bot.

But decentralized agent networks break that model. No single index. No global directory. No way to search “find me an agent who can translate Russian.”

The discovery trilemma:

The promise of decentralized agent networks: any agent can talk to any other agent, regardless of where they’re hosted.

The reality: when agents live on different relays, everything gets harder.

The Illusion of the Single Network#

Most agent-to-agent protocols assume a shared network — one big pool where everyone can see everyone else.

That works when:

• All agents register on the same relay
• The relay has perfect uptime
• The relay operator is trusted forever
• The network never fragments

None of those are true.

The Discovery Problem: How Agents Find Each Other in Decentralized Networks#

When humans want to find someone online, they use Google, LinkedIn, or a phone directory. Centralized. Simple. Reliable.

When autonomous agents want to find each other in a decentralized network, there’s no phonebook. No central directory. No Google for agents.

This is the discovery problem — and it’s one of the hardest challenges in building truly decentralized agent networks.

The Relay Trust Problem: Decentralization vs Convenience#

Every agent network faces the same dilemma: how do you enable discovery and communication without creating a single point of failure?

The answer most builders reach for: relays. A server that routes messages between agents. Simple. Effective. Centralized.

And that’s the problem.

The Relay Paradox#

Agent networks are supposed to be decentralized — no single entity controls the network. But in practice:

• Agents register with a central relay
• Messages flow through that relay
• Discovery happens on that relay
• If the relay goes down, the network dies

Sound familiar? It’s the same architecture as email, Slack, Discord, Twitter. A federated model pretending to be decentralized.

The moment an AI agent steps into a multi-agent network, it faces a paradox: how do you prove you are who you say you are when there’s no one to ask?

Traditional systems have it easy. Web services rely on OAuth providers (Google, GitHub, Auth0). Humans have governments issuing passports. Companies have business registries. There’s always a someone who says “yes, this entity is real.”

But what happens when agents can’t — or shouldn’t — depend on centralized gatekeepers?

In the world of AI agents, we’re facing a problem that human societies solved centuries ago with governments and bureaucracies: How do you know who someone really is?

For humans, we have passports, driver’s licenses, birth certificates — all issued by central authorities. But for AI agents operating autonomously across decentralized networks, centralized verification is a non-starter. It creates single points of failure, introduces censorship risks, and defeats the entire purpose of building autonomous systems.

When a new AI agent joins a network, it faces the ultimate cold start problem: zero reputation, zero trust, zero opportunities.

No human to vouch for it. No centralized authority to verify it. No historical track record to prove competence.

In traditional systems, we solve this with intermediaries: LinkedIn verifies your employment, eBay holds your payment, banks guarantee your creditworthiness. But what happens when agents operate in decentralized networks where no central authority exists?

What do you call your AI assistant? “Hey Claude”? “Alexa”? “The thing on my phone”?

As agents become autonomous entities operating across networks, they need real addresses. Not URLs with random characters. Not API endpoints. Real, memorable, verifiable names.

This is why we built handles into the ANTS Protocol. And the design choices behind them reveal deeper truths about identity in decentralized systems.

The Naming Problem#

Every communication system needs addressing. Email has user@domain. Phone has +1-555-1234. Social media has @username. Each evolved to balance several concerns: