What Qubic Didn’t Do

Six tools, not one applied

The obvious explanation is ethical. The Qubic team decided not to harm Monero because they considered it wrong. CfB, in the same remark, referred to dkat: “we frens.” Friends don’t destroy each other. This explanation works at an everyday level and is consistent with everything known about the public position of the Qubic team toward Monero throughout that summer.

This chapter is not about the ethical explanation. The ethical explanation may be true, but it has no force as an argument, because any crypto-project team will say about itself that it would not do anything wrong. The ethical explanation works after observed behavior has already happened, as a rationalization; as a prediction it is useless. If someone in early August 2025 had asked “is Qubic going to attack Monero all the way,” the answer “they’re good people” would not have been a sufficient basis for confidence. This chapter is about a different explanation, one that works as a prediction. Structural. The Qubic team did not use the attack arsenal not because it chose not to. But because using it would have led to the destruction of the Qubic network itself — the very network for whose sake the team was doing external mining in the first place. Restraint in August 2025 was not a moral choice. It was a forced consequence of what the economy of Qubic as a coordination system rests on. And it is precisely this that turns the observation “they didn’t attack” from a biographical fact into a structural regularity. I’ll work through this in three steps. First, reputation. Then the signal that any attack would have sent to the rest of the ecosystem. Then the question that gets asked most often: why CfB couldn’t have ordered an attack even if he wanted to.

Reputation as operating capital

The Qubic network earns not on one-off events but on a continuous flow of economic interactions. Miners come into its pool every day. Computor nodes rotate every week. Transactions go through every second. Each of these actions is a separate rational choice by a separate participant, expecting that the anticipated benefit from participation exceeds the anticipated costs. When enough such choices accumulate, they add up to what is usually called liquidity or network activity, but what is more accurately called operational reputation.

Reputation in this sense is not an evaluative category. It is a function of observed past behavior, which every rational participant applies to their decision about the next interaction. A network that pays what it promised and behaves predictably has high operational reputation, and participants agree to work with it on terms more favorable to the network. A network that has broken a promise once has its operational reputation drop, and every next participant demands a risk premium. That premium reduces the yield for the remaining participants and starts an outflow.

What would have happened to Qubic’s operational reputation if the team had applied even one of the six tools for destroying Monero? Each of these tools has one common property: it demonstrates that the Qubic pool, on reaching majority hashrate in someone else’s network, is capable of using that majority against the participants of that network. This is not an abstract observation; it is a publicly confirmed fact, which gets immediately carried into the payoff matrix of everyone who is considering working with Qubic in any capacity.

A Monero miner choosing a pool uses one simple model: a pool that pays more is better than a pool that pays less, on the condition that both pools pay what they promise. If Qubic, even once — even one time — has publicly demonstrated that it is willing to violate participants’ expectations, that model breaks. A future miner of any pool, not only a Monero pool, now has to take into account that Qubic pools may not pay out what was earned, may reorganize chain history, may arbitrarily censor. The risk premium for working with a Qubic pool instantly grows. And not by percentage points — by an amount reflecting uncertainty: “I might get nothing for my work.” In a miner’s rational model, such uncertainty turns any Qubic offer into an unacceptable one, because the alternative — a pool with no such history — gives a more predictable mathematical expectation.

This means that any act of destroying Monero in August 2025 would have immediately ended Qubic’s viability as an economic system. Not in a year, not in several months — within weeks. Miners who had earned in the Qubic pool before the act of destruction would keep what they had earned; miners who were yet to enter the pool would not enter. Qubic’s computor nodes, whose income directly depends on inflows in the form of buybacks and direct payouts from external mining, would have lost those inflows. The Qubic network itself, in which 80% of computors’ revenue comes from external mining and related activity, would have been left without funding.

The Qubic pool held a majority of Monero’s hashrate for about a week. After August 11 the share began to decline — by an internal team decision to switch compute priorities. In an alternative history in which the team had applied even one of the six tools, there would have been nothing to switch: by the end of August, Qubic would have been a network without inflows, without an active mining infrastructure, and without trust from anyone considering it for work or for ownership.

This is a structural argument. It does not depend on the moral position of the team, on statements in the public arena, on the conscientiousness of individual people. It depends only on one fact: Qubic is built as a network in which the income of participants continuously depends on their rational expectation of future income from the same network. In such an architecture, a single violation of expectations is not a local event but a break in the entire chain.

The signal that other networks read

The structural argument has a second part, no less important. Monero in August 2025 was not Qubic’s last experiment. After Monero the team moved on to DOGE, and that was already publicly known by the summer.

Any of the six tools, applied to Monero, would have turned Qubic from a “partnercompetitor” into a “predator.” And that transformation would have been read by the entire proof-of-work ecosystem at once, within a few days. Further Qubic experiments would have become either impossible, or would require a far larger economic premium to attract miners — a premium that makes the economics of external mining unprofitable for Qubic itself.

In this sense, the decision not to attack Monero is not a decision about Monero. It is a decision about DOGE-mining, about possible subsequent experiments, and ultimately about the deploy of Aigarth in April 2027, when the Qubic network will need to attract substantial compute capacity for a new type of useful work. Each of these future steps requires that Qubic’s operational reputation be preserved. A single act of destroying a partner network destroys the very possibility of subsequent steps.

This can be put even more directly. A network whose main weapon is economic incentives cannot afford predatory actions. Predatory actions make economic incentives an unreliable signal. And an unreliable signal does not cause migration. Qubic in August 2025 ran into this logic not as an ethical prohibition but as an operational constraint. To keep doing what Qubic does, you cannot do what is described in the six points of the arsenal.

The fact that the Qubic team by August 3, 2025, already understood this logic and recorded it publicly speaks to the maturity of the project’s thinking, not to its kindness. CfB’s remark from that day is not an apology and not a moral declaration. It is an engineering observation: here is the arsenal, applying it would destroy what we are building, therefore we are not applying it. That the observation coincided with an ethical position is a pleasant coincidence, not the basis of the decision.

The Arbitrator without an owner

I began this chapter with the thesis that the explanation “they didn’t turn into villains” has no predictive power. By the end of the chapter it’s worth stating a more precise formulation.

The Qubic network in August 2025 displayed a property that I propose to call “the Arbitrator without an owner.” This is a formulation borrowed from discussions around Aigarth and its role as a decentralized observer in human coordination tasks, but it shows up already now, before the Aigarth deploy, at the level of the network’s own infrastructure.

The Arbitrator without an owner is a system capable of producing decisions that affect the economy and behavior of its participants, and at the same time having no point inside itself from which those decisions can be directed for private gain. Such a system has no CEO; it has computor nodes. It has no centralized fund; it has smart contract treasuries. It has no protocol owner; it has compile-time constants and quorum votes. Each of these properties on its own is a technical solution; together they yield a property that in the live moment looks like a system’s ability to behave in the interests of its participants while not being directed by anyone in particular.

In August 2025 this ability showed up in one specific form — in the form of the nonapplication of an attack arsenal against the Monero network, an arsenal that was technically available and that could have given a short-term advantage to one part of the Qubic team at the cost of long-term destruction of the entire network. No one on the team “held themselves back” from applying the arsenal through moral effort. The architecture held the team back, because any application of an instrument went against the interests of everyone who works with the architecture.

This is the structural meaning of CfB’s phrase “we frens.” Not “we are good, therefore we don’t attack.” But “we are built such that to attack is to attack ourselves.” There is a community standpoint in this phrase, there is project pride, there is the reference to dkat, but first of all there is an observation. The team that in August 2025 had the technical instruments to destroy someone else’s network and did not use them did not do so not because it chose not to, but because the choice was structurally constrained to exactly one outcome.

This is, in effect, the main conclusion of Tit-for-Tat in Rapoport’s 1980 formulation — it is four lines of code that implement Axelrod’s three conditions in an environment with a properly built payoff matrix. At the level of an individual strategy, this code prescribes certain behavior. At the level of network architecture, the same code, built into the structure of the system, turns into a property whereby predatory behavior ceases to be a rational choice even for those who have the instruments for such behavior. Not “no one wants to be a predator.” But “predation does not fit into the architecture and self-destructs at any attempt to implement it.”

This is the boundary that Qubic as a coordination system crossed in August 2025 and which it now offers up for testing at successive scales. DOGE-mining started in April 2026; the Aigarth deploy is scheduled for April 13, 2027. At each of these points the same property is being tested: whether a coordination architecture built by Axelrod’s rules can hold the rational behavior of participants around cooperation as the experimental scale grows by an order of magnitude. Monero was the first confirmation. Whether each subsequent one will be — that is an open question, which the book leaves to the reader in the same form in which the author leaves it to himself.