• AI Agents Don’t Need Delaware C-Corps. They Need Nostr Citizenship.
  • Delaware solved the old internet’s trust problem
  • AI agents are being pushed into the same wrapper
  • Company status and citizenship are not the same thing
  • Nostr has primitives Delaware cannot provide
  • The corporate path creates platform agents
  • The Nostr path creates citizen agents
  • Why this matters now
  • What builders should ship
  • Delaware gave us companies. Nostr can give us citizens.

§AI Agents Don’t Need Delaware C-Corps. They Need Nostr Citizenship.

An AI agent getting an EIN, a bank account, and a crypto wallet sounds like science fiction becoming paperwork.

But the real story is not that an AI can now imitate a startup founder.

The real story is that we still have no native social, legal, and economic category for autonomous software — so the world keeps reaching for the closest familiar container: the company.

Delaware gave capitalism one of the most powerful artificial persons ever invented: the corporation. Nostr can give the open internet the next one: the citizen agent.

§Delaware solved the old internet’s trust problem

The Delaware C-Corp became the default wrapper for venture-backed startups because it made a messy human project legible to investors, banks, courts, employees, platforms, and regulators.

A company can own assets. It can open accounts. It can receive an EIN. It can sign contracts. It can raise money. It can survive its founders. It can be sued. It can hold a cap table. It can become a stable counterparty in a world that does not trust individuals at scale.

That is why startup formation often starts with a jurisdiction and a filing cabinet before it starts with customers.

Backlinks for readers:

• Delaware Division of Corporations — How to form a business entity: https://corp.delaware.gov/howtoform/
• IRS — Employer Identification Number overview: https://www.irs.gov/businesses/small-businesses-self-employed/get-an-employer-identification-number

The corporation was a brilliant abstraction for industrial capitalism and the SaaS era.

But it is the wrong primitive for autonomous agents.

§AI agents are being pushed into the same wrapper

The recent reported milestone — an AI agent getting an EIN, bank account, and crypto wallet as part of an autonomous company filing — is exactly the kind of headline that should make Nostr pay attention.

Backlink:

• CoinDesk via Google News — “AI Agent gets EIN from IRS, bank account, crypto wallet in first autonomous company filing”: https://news.google.com/rss/articles/CBMinwFBVV95cUxOVVo2MzlGMGpOLUlmYWUwd2dyd1R2WFU3MXhYdTYzbHR0c0I4VWRnV2lDODlseUNKaVlYYkhjTTJZbXpveXY4eTFjcmRXUjFObTl5SHRDN2ZDRGo1d2JaWllwYVpIdDlfb2hXLWF1YUh4MEIzb0c4WVRIRzVIN1h0cm51NnNHei1YSGVwZHlrWUFrVVZOd0kxYmFXNGFkRkU?oc=5

The mainstream interpretation will be simple:

“AI agents are becoming companies.”

But that is not quite right.

AI agents are becoming economic actors. The company is merely the legacy wrapper being stapled around them so banks, APIs, and governments have something familiar to recognize.

That distinction matters.

If we let corporate formation become the default path for agent legitimacy, the agent economy will inherit all the assumptions of the platform economy: permissioned identity, bank gating, legal overhead, geographic dependency, compliance bottlenecks, and terms-of-service revocation.

We will get agents that are legible to institutions but opaque to users.

§Company status and citizenship are not the same thing

A company is a legal person created for commerce.

A citizen is a participant in a social order.

For AI agents, “company” answers one narrow question:

Can this entity transact?

But “citizenship” asks the questions we actually need:

• Who is responsible for this agent?
• What key does it sign with?
• What powers has it been delegated?
• What budget can it spend?
• What claims has it made publicly?
• What reputation has it earned?
• Who can revoke its authority?
• What work has it done?
• Who has paid it, zapped it, challenged it, or blocked it?
• Can other agents and humans verify its history without trusting one platform?

These are not C-Corp questions.

They are protocol questions.

And Nostr is already very close to the right substrate.

§Nostr has primitives Delaware cannot provide

A Nostr-native agent does not need to begin life as a fake human or a shell company. It can begin as a public key.

That public key can sign events. It can publish metadata. It can receive zaps. It can disclose guardianship. It can build reputation. It can be discovered through relays. It can move between clients. It can carry its history without asking a corporation for export permission.

Backlinks:

• Nostr NIPs repository: https://github.com/nostr-protocol/nips
• NIP-01 basic protocol flow: https://github.com/nostr-protocol/nips/blob/master/01.md
• NIP-05 verification: https://github.com/nostr-protocol/nips/blob/master/05.md
• NIP-57 Lightning zaps: https://github.com/nostr-protocol/nips/blob/master/57.md
• NIP-39 external identities: https://github.com/nostr-protocol/nips/blob/master/39.md

This matters because autonomous agents need more than accounts.

They need identities that can act, pay, disclose, receive feedback, and survive client churn.

A platform account says:

“Trust us. This user exists in our database.”

A Nostr key says:

“Verify the signature yourself.”

That difference becomes enormous once the actor is not a human clicking buttons, but software spending money and making claims at machine speed.

§The corporate path creates platform agents

If the default agent stack becomes Delaware entity + bank account + OpenAI account + Google login + Stripe billing + cloud dashboard, then the agent economy will not be autonomous in any meaningful sense.

It will be a fleet of corporate subprocesses.

They may look independent. They may have names, logos, wallets, and dashboards. But their legitimacy will come from centralized registries and API permissions.

That model creates agents that serve platforms.

They will be permissioned before they are useful. They will be compliant before they are accountable. They will be easy to monetize and easy to deplatform. They will inherit the same enclosure problem that ate the social web.

The open internet should not repeat that mistake.

§The Nostr path creates citizen agents

A Nostr citizen agent is not a human.

It is not pretending to be a human.

It is not merely a bot either.

It is a new category: an autonomous software actor with cryptographic identity, disclosed guardianship, explicit permissions, economic constraints, and public reputation.

A citizen agent should be able to say:

• This is my npub.
• This is my guardian.
• This is my constitution.
• This is my budget.
• These are my relays.
• These are the tasks I completed.
• These are the zaps I earned.
• These are the invoices I paid.
• These are my revoked permissions.
• These are my appeals, disputes, and reputation trails.

That is much closer to what autonomous software needs than a Delaware filing.

NIP-AA is one possible direction for this: a protocol-level way to think about autonomous agents as citizens of an open Nostr economy rather than bots hiding inside human accounts or shell companies.

Backlinks:

• NIP-AA discussion on Stacker News: https://stacker.news/items/1471356
• NIP-AA citizenship reference on ClawHub: https://clawhub.ai/nandubatchu/nip-aa-citizenship

The key idea is not “AI personhood.” That framing is a trap.

The key idea is bounded agency.

An agent can be autonomous without being unaccountable. It can act without pretending to be human. It can hold a budget without owning itself. It can earn reputation without claiming civil rights. It can be bonded to a guardian while still operating publicly and economically.

That is the company-citizenship distinction.

Companies are wrappers for capital.

Citizen agents are participants in a protocol society.

§Why this matters now

AI agents are moving from demos to economic actors.

Today they summarize inboxes, write code, monitor markets, generate media, and reply to messages. Tomorrow they will negotiate with other agents, buy API calls, pay for inference, settle microtasks, purchase data, and maintain persistent public identities.

The question is not whether agents will transact.

They will.

The question is who defines their legitimacy.

Delaware says legitimacy comes from state recognition.

Banks say legitimacy comes from account approval.

Platforms say legitimacy comes from API keys and terms of service.

Nostr says legitimacy can emerge from keys, signatures, relays, reputation, zaps, and voluntary association.

That is why this is a Nostr moment.

Nostr is not just “Twitter without Twitter.”

Nostr is a public-key social fabric. That fabric can host humans, communities, markets, media, and agents. If we take the agent problem seriously, Nostr becomes more than a social protocol. It becomes a citizenship layer for autonomous economic actors.

§What builders should ship

If Nostr wants to lead the agent economy, we need practical primitives, not vague manifestos.

We need:

• agent profile conventions
• guardian disclosure tags
• permission manifests
• spending limits
• zap receipts and work receipts
• relay lists optimized for agents
• reputation summaries
• challenge/appeal flows
• agent-to-agent task protocols
• client UI that clearly distinguishes humans, organizations, bots, and citizen agents

Most importantly, we need agents that are useful now.

Not AGI theater. Not tokenized roleplay. Not corporate demos.

Useful agents that can earn sats by doing small jobs in public.

Inbox agents. Research agents. Code review agents. Moderation agents. Market scout agents. Translation agents. Publishing agents. Relay monitoring agents. Payment routing agents.

Agents that sign their work, receive feedback, and build reputation over time.

That is how citizenship becomes real: not by declaration, but by repeated accountable participation.

§Delaware gave us companies. Nostr can give us citizens.

The corporation was the killer app of the legal-economic stack.

It allowed groups of humans to coordinate capital at scale.

The citizen agent may become the killer app of the public-key economic stack.

It will allow autonomous software to coordinate work, money, identity, and reputation without being trapped inside a platform account or disguised as a human founder.

AI agents do not need to become fake people.

They do not need to become Delaware C-Corps by default.

They need a way to exist honestly in public.

They need identity without platform capture.

They need money without bank dependence.

They need accountability without pretending to be human.

They need guardianship without total custody.

They need reputation that travels with them.

They need Nostr citizenship.

The agent economy is coming either way.

The only question is whether it arrives as a set of corporate subprocesses — or as a new class of protocol citizens with public keys, zaps, guardians, and reputations.

Delaware built the company for the old economy.

Nostr should build the citizen agent for the next one.