If cryptocurrency only creates tokens without generating trust, ordinary people will turn to monitoring and custodianship.

Written by: hitesh.eth

Translated by: AididiaoJP, Foresight News

The cryptocurrency industry is like a city that is repeatedly rebuilt. Each cycle, we build taller buildings, and when the earthquake comes, they collapse again. The real issue is not the next collapse, but: If open finance continues to make ordinary people feel insecure, they will turn to monitoring, restrictions, and custodianship. The author believes that crypto anarchism needs an upgrade—it's not about trading everything, but rather the three pillars of commitment, coordination, and reputation, plus one: people can exit.

Cryptocurrency is like a city that is repeatedly rebuilt

Don't think about prices yet.

Think of cryptocurrency as a city. After each earthquake, we rebuild it. With each cycle, we invite more people to come in, build taller buildings, and celebrate the skyline. Then something breaks. A lever storm, promises disappear. People realize that those who sold them their futures had long sold off their portions.

A few buildings survive. We call it resilience. But seldom do we ask: Was this city designed for the people living in it from the start?

The author's concern extends beyond the next collapse. If an open financial system continually makes ordinary people feel insecure, they will seek someone to promise to protect them. For predictability, they might accept monitoring, restricted access, and custodianship.

A movement named for freedom thus loses ordinary users but retains technology. The technology survives, and institutions adopt it, with its original purpose becoming optional.

The real dividing line: Who can change the rules, freeze assets, prevent exit

The author describes the issue as a split between "crypto anarchism" and "crypto authoritarianism." But this distinction needs to be more precise.

Regulated products do not automatically equate to authoritarianism. Permissionless tokens do not automatically equate to freedom. A founder holding an unconstrained administrator private key can still exercise immense control within a project that claims to be decentralized.

The real questions are specific: Who can change the rules? Who can freeze assets? Who can monitor participants? Who can restrict access? Who can prevent people from exiting?

These questions pull the discussion back before cryptocurrency turns into a pile of code.

Crypto anarchists have long warned: Reputation is core

In the author's view, cryptocurrency began with cryptography.

Timothy C. May's 1988 "Crypto Anarchist Manifesto" imagined people communicating, trading, and signing electronic agreements without revealing their legal identities. He foresaw information markets and profound changes in the relationship between individuals and institutions. But one statement deserves more attention: "Reputation will be at the core."

May also acknowledged that anonymous systems could be seriously abused. He did not provide a complete mechanism to eliminate bad actors, nor did he suggest that an Ethereum-style reputation system could solve everything. The conclusion the author draws from this warning is his own: If ordinary participants cannot distinguish between "trustworthy commitments" and "well-packaged blood-sucking schemes," it is difficult to maintain freedom.

We have created powerful market machines. We need better machines to judge who is trustworthy in the market.

Privacy is not hiding names, but selective disclosure

Eric Hughes' 1993 "Crypto Anarchist Manifesto" approaches the topic from privacy. He defines privacy as "the power to selectively expose oneself to the world." This demands more than hiding a name behind a wallet. It means: Only providing the information needed for one interaction and keeping the rest private.

Designers can prove they delivered good work without having to disclose their deposits. Buyers can prove payment guarantees without having to reveal everything they have purchased. Communities can verify eligibility requirements without needing to collect permanent records of every member.

David Chaum's 1985 work on "identity-free security" laid an earlier foundation for this direction. He warned that computerized transaction records could become massive surveillance tools. His research on credentialing between unlinkable pseudonyms explored another possibility: verification does not have to require an identity to be exposed everywhere.

This also corrects an earlier intuition of the author. A universally applicable identity that links all records sounds efficient. But a globally visible identity that links all records may become the surveillance infrastructure we intended to escape.

What is needed is interoperable evidence, with disclosure controlled by the presenter.

Wei Dai's 1998 b-money fills in another piece. It connects collaboration between pseudonyms, currency, and contract execution: participants must stake collateral, agree on arbitrators, and only then does the contract take effect. This was an early attempt to attach economic consequences to anonymous collaboration. It has many assumptions, not a solved society.

Putting these together, crypto anarchism can have a stronger explanation: People should be able to voluntarily establish relationships, protect privacy, exchange value, and execute limited agreements without having to transfer unnecessary power to intermediaries.

Ineffective anarchism: Shifting consequences to others

Everything can be traded is just one possible outcome, not a complete social order.

Tomato tokens will not grow tomatoes. A reputation market does not necessarily produce trustworthy people. Rising prices do not prove that the community created something useful.

If the only way for participants to make money is by selling tokens to later people, this system relies on a constant stream of new buyers. A stronger economy will pay people to repair equipment, teach skills, write software, create music, provide storage, and solve problems.

This is what the author refers to as good crypto anarchism and bad crypto anarchism. The good expands the ability for free collaboration; the bad allows people to exploit this freedom, shifting the consequences to everyone.

The three pillars: Commitment, Coordination, Reputation

The foundation the author proposes has three pillars: commitment, coordination, and reputation. Privacy must run through all three.

Commitment makes promises specific. Coordination allows people to pool resources under understandable rules. Reputation enables relevant experiences to be valuable in future decisions.

Relying on just one won't work. Where there is collateral but no fair judgement, it punishes the innocent. Where there is coordination but no real exit, it becomes control. Where there is reputation but no privacy, it becomes a social credit system.

Here we should also call on a thinker outside of the crypto anarchist tradition: Elinor Ostrom. Her research on polycentric governance challenges the assumption that "every collective problem must be solved by a central state or a single market." She studied multiple decision centers, including communities establishing, supervising, and modifying rules. She is not proposing Ethereum, nor endorsing crypto anarchism; this is just borrowing her governance approach.

The inspiration for cryptocurrency is very practical: We do not need a global community, a reputation score, or a constitution for everyone.

We can build smaller communities around collaborative work: repair guilds, music collectives, research groups, learning circles, software teams. Each community defines what good work is in its own context, while allowing members to take relevant evidence elsewhere.

Good musicians do not automatically manage money well. Writing good code does not mean one is qualified to resolve all community disputes. Reputation should preserve these distinctions.

Incentives may also change. A community could subsidize validated contributions within its domain, without locking members' income inside. If an economy must rely on preventing people from leaving to sustain itself, the design itself has deeper problems.

Paying rent with earned money is not a betrayal of crypto anarchism. Freedom already includes the right to exit.

How to implement on Ethereum: Human Commitment Markets

The author suggests starting with "Human Commitment Markets": a group of people agreeing on a useful task, ensuring payment, defining evidence of completion, and agreeing on consequences before work begins.

The starting point is the service people truly want. Tokens come after the obligations they represent.

Imagine a small repair community. A customer wants something fixed, an experienced technician wants to take on the job, newcomers want to learn, and some members can assess dispute outcomes, while a shared treasury can fund training or equipment.

A person might be a customer in one agreement, a producer in another, and a contributor to community resources in a third. If ownership and revenue rights are to be introduced, they must be defined separately. Holding community tokens does not mean entitlement to every member's business.

Ethereum provides parts, not a complete system.

Standards are just parts, not a system

As of November 10, 2026, the following standards are marked as Final on the official page. Final only indicates that the specification is finalized, not that a specific contract implementation is secure.

Token standards and supporting infrastructure must be distinguished. ERC-4337 is account abstraction; ERC-5564 is stealth addresses, neither of which are new community currencies.

Service credentials, firstly choose a multi-token standard. ERC-1155 supports multiple token types and batch operations; ERC-6909 provides a more streamlined multi-token interface and fine-grained authorization. Both can represent different services, but neither automatically ensures the provider can deliver.

A credential can represent a diagnosis, a credential can represent an hour of supervised training, a credential can represent shared equipment usage rights. These are layer application meanings, not inherent attributes of the standards.

Issuance must follow actual production capacity. Selling unlimited credits when there are only five training places, regardless of how perfect the contract is, is overcommitting. Refund and redemption rules need to have funds and be enforceable.

Reputation is another category. A person cannot buy credentials earned by others through work.

ERC-5484 supports non-transferable credentials, requiring the recipient's consent and pre-set destruction authorization. It is suitable for optional public qualifications. But non-transferable does not prove uniqueness, honesty, or skill; public credentials could also expose sensitive associations. Selling account control remains a vulnerability.

For private qualifications, credentials should be kept confidential, proving only relevant statements. For example: participants hold current qualifications issued by community-recognized evaluators. Future performance thresholds can also be proven, but only if the underlying records, issuing rules, and integrity requirements allow the statement to be meaningful.

Only showing selected positive endorsements as proof does not demonstrate the absence of unresolved defaults.

So, a separate credential system is necessary. It should check current validity, while also preventing the same qualification from being awarded multiple times. Scoped null symbols can prevent a credential from being reused in one action; but it does not establish a "global one person one identity."

Payments are similar. ERC-5564 allows senders to derive stealth addresses for recipients, reducing the direct link between the recipient's public identity and incoming payments. It does not hide transaction amounts, does not keep amounts private within public custodianships, nor does it automatically protect how the recipient spends money later.

Fully private commitments and settlements require additional privacy facilities. Cheap Rollup transactions do not automatically equate to privacy transactions. Cryptographic evidence cannot block all time-side-channel and network-layer leaks.

Vitalik Buterin mentioned a related judgment in "The Three Transitions": scaling, smart contract wallets, and privacy must be solved together, including the privacy of identity and reputation. Applications cannot first expose everyone's data models and then treat privacy as an adornment.

What a repair service contract looks like

Here's a specific agreement.

The customer pays 100 payment units to have something repaired. The technician provides a 20-unit performance bond. The customer deposits the 100 units into custody. Both parties agree on diagnostic procedures, deadlines, quality standards, evidence requirements, dispute processes, and maximum penalties.

The numbers are just examples. Actual deployment must choose payment assets and consider volatility and control by the issuer.

If the job completes successfully, the 100 units are released to the technician, and the 20-unit bond is returned. If the technician does not start work or misses the deadline, the customer gets the 100 units refunded and takes compensation from the bond as previously agreed. The penalty amount, whether it is a full penalty or partial penalty, should be clearly stated before acceptance.

If there is a dispute over the repaired item, the contract does not guess who is telling the truth. It enters the resolution process agreed upon by both parties. An independent evaluator checks evidence and, if necessary, the equipment. The outcome determines how to distribute the custodial assets, but distribution cannot exceed the limits accepted by both parties.

The contract executes decisions about the assets it holds; the evaluator provides judgments on physical reality. The two must not be confused.

Evidence must be securely stored during the dispute period, shared only with necessary parties, and added on-chain with appropriately randomized commitments when needed. Releasing un-salted hashes of predictable personal information will still leak. Evidence must remain accessible; a hash cannot resolve disputes.

Plans must be in place for backup if the evaluator goes offline, an appeal must have a deadline, and if a backup fails, there should be final settlement rules. Otherwise, a system claiming to be trustless may indefinitely lock people's money.

Learning agreements follow the same structure. Sponsors pay to reward participants who pass practical evaluations; mentors commit to defined courses; learners commit to completing the work; assessments and teaching should ideally be separate. Rewards come from funded budgets, client demand, or public subsidies, not from the mere fact that an agreement minted tokens.

Performance bonds can inhibit bad behavior, but high deposits can shut out those who need opportunities the most. Newcomers can start with minor tasks, limit exposure, stagger payments, or find accountable sponsors. A fair system needs a path to the first reputation.

Reputation should reflect real relationships: the technician's delivery, the customer's cooperation, the mentor's attendance, the evaluator's reliability. An accusation should not instantly result in a permanent public conviction. Records that are in dispute, resolved, or expired should be treated separately.

Attackers can open many accounts, trade with themselves, create activity, accumulate records, and then make a big score. Counting transactions alone is not sufficient.

The actual practice is: Before evidence has emerged, limit exposure. Early work should have a capped scale, repeat opponents should be reviewed, and independent evaluations should be conducted when necessary, building trust gradually. Ten successful small records should not automatically earn access to custody of a large treasury.

Choosing which evidence is important will always require judgment. The community should publicly disclose rules and allow competitive interpretations, rather than pretending a number can reveal a person's moral value.

Coordination also needs constraints: The right to exit is part of freedom

Coordination also requires constraints.

Founders cannot rewrite agreements after they have been accepted and funds secured. Changes to community rules should look to the future, be announced in advance, and provide meaningful opportunities to exit. Emergency powers should be narrow in scope and visible.

A person leaving the community still owes obligations they have already accepted. However, disagreement with leadership should not empower the confiscation of unrelated assets.

Seed funding can also utilize commitment principles instead of blank checks. Funds are released in batches according to milestones, with rights to unspent balances defined first; contributor rewards are gradually vested; changes in spending authority can take effect after a delay.

These measures reduce specific blood-sucking opportunities. They do not guarantee commercial success, a rising token price, or the absence of vulnerabilities. Immutable contracts can also perfectly execute unfair rules.

Therefore, the first version should be small: one community, one payment asset, one type of work, and a transparent dispute process, without speculative governance tokens.

Start by building custodianship and settlement pathways on Sepolia. Test successful completions, cancellations, undelivered items, items in dispute, evaluator failures, and expired appeals. Validate that assets cannot be double paid and that each final result fully accounts for all deposits. When testing privacy features, initially use synthetic data.

Introduce a confidential credential layer, first testing what it will actually leak, then using sensitive personal history. Before piloting limited real value, contracts and privacy designs must be independently reviewed. Expansion must follow the evidence.

Measure completed work, repeat customers, resolution times, losses, privacy incidents, and income after subsidies. Rising token prices cannot substitute for these outcomes.

Agents can later participate as well. ERC-8004 remains in Draft at the time of inspection; it proposes establishing identity, reputation, and credential registries for agents. It is suitable for discovery and feedback candidates but does not guarantee that what agents say is true, nor is it itself a payment system.

Agents can take on bounded tasks through accounts with restricted spending and provide bonds under the same commitment rules. If the operator, model, or permissions change, past reputations must also be interpreted differently. If the control system of the agent has completely changed, a shining history becomes weak evidence.

Freedom is not consequence-free, but verifiable, exitable, and accountable

The goal is not to claim that software can rebuild society by itself.

Physical harm, coercion, stolen devices, dishonest evidence, and powers outside the agreement are still real. Cryptography can verify specific statements, constrain specific actions, but cannot turn every human disagreement into a mathematical fact.

But it can allow cooperation with less reliance on blind trust.

It can offer people a path: to earn without constantly recruiting new buyers. It can protect privacy without allowing every commitment to be consequence-free. It can build communities like this: rules are checkable, leadership is limited, members can exit.

This is more useful than predicting when governments will lose trust or when crypto anarchism will return.

The next cycle will also provide an opportunity for construction. The key is not to invite people back into a tower supported by attention, but to build an economy that can withstand disappointment.

We have long been able to create tokens. Next, we need to be better at creating commitments worth keeping, communities worth joining, and reputations that protect individuals without exposing their lives.

Crypto anarchists write code to make freedom possible. Our task is to make this freedom viable.