One provocative theory suggests that Bitcoin could be attacked and shorted at the same time. The idea is unsettling—but turning it into reality would require far more than a large bank account.

For years, Bitcoin’s security has rested on a powerful economic assumption: attacking the network should cost more than any attacker could reasonably hope to earn.

Campbell Harvey, a finance professor at Duke University, recently challenged that assumption. During an episode of The Wolf of All Streets, he outlined a disturbing scenario. A sufficiently wealthy organization could spend roughly $8 billion to obtain enough computing power to disrupt Bitcoin while simultaneously opening enormous short positions against BTC.

If the attack triggered a market collapse, the profits from those positions could theoretically cover the cost of the operation—and perhaps leave the attacker with billions in additional gains.

It is a clever thought experiment. It also sounds like the plot of a financial thriller.

But could it actually work?

The Attack That Pays for Itself

Bitcoin uses Proof of Work to determine which version of its transaction history the network should accept. Miners compete to produce valid blocks, and nodes generally follow the valid chain containing the greatest amount of accumulated computational work.

An attacker controlling a majority of the active hashrate could temporarily produce blocks faster than the rest of the network. This is commonly described as a 51% attack.

Such control would be serious, but it would not grant unlimited power. According to Bitcoin’s developer documentation, a majority attacker could reorganize recent transactions or stop new ones from confirming. The attacker might also reverse their own payments and spend the same coins twice.

However, they could not generate valid signatures for someone else’s wallet, confiscate arbitrary coins or unilaterally increase Bitcoin’s 21-million supply. In other words, controlling mining does not mean controlling every rule of the protocol.

The objective would therefore be psychological as much as technical. Repeated reorganizations, double-spends or prolonged censorship could convince exchanges, institutions and ordinary holders that Bitcoin was no longer reliable. The resulting panic could send its price sharply lower.

That is where derivatives enter the equation.

Harvey’s argument is that modern futures, options and perpetual-swap markets have created a potential revenue source that did not exist at Bitcoin’s birth. An attacker would no longer need to make money from the blockchain itself. They could profit from the financial damage occurring around it.

The widely repeated $8 billion figure is based on an estimate placing the required mining equipment at approximately 0.5% of Bitcoin’s value when its market capitalization was around $1.6 trillion. But this is a rough calculation, not a genuine industrial quotation. It assumes hardware can be purchased at something close to its current market price and largely ignores the cost of building the infrastructure around it.

That is where the elegant theory begins to collide with the physical world.

Bitcoin Cannot Be Attacked From a Spreadsheet

Purchasing billions of dollars in mining machines is not comparable to buying a liquid stock. Bitcoin ASICs are specialized computers produced by a relatively small number of manufacturers. Global inventories, semiconductor capacity and delivery schedules are all limited.

As an illustration, Bitmain’s Antminer S21 specification lists 200 terahashes per second of computing power while drawing approximately 3,500 watts. Matching a Bitcoin network measured in hundreds of exahashes per second would require several million machines of this class.

Those machines would also need somewhere to operate.

An attacker would require mining facilities, transformers, substations, cooling equipment, networking infrastructure, technicians, replacement parts and access to an extraordinary amount of continuous electricity. The challenge would not simply be paying the energy bill. It would be securing power capacity comparable to that consumed by the existing global mining industry.

A 2025 Cambridge study estimated Bitcoin mining’s annual electricity use at approximately 138 TWh. Reproducing a large share of that infrastructure in secret would be closer to launching a multinational industrial sector than setting up a data center.

The market would also react.

If one buyer began absorbing a substantial portion of the world’s ASIC supply, manufacturers would raise prices, delivery times would grow and competitors would notice. Existing miners could invest in additional machines, forcing the attacker to chase a moving target. Regulators, energy companies and intelligence agencies might also become interested long before the operation was ready.

Mining-pool concentration does not offer an easy shortcut either. A mining pool may coordinate substantial hashrate without owning the underlying equipment. Independent miners can redirect their machines to another pool if an operator begins behaving maliciously.

A state could theoretically seize existing farms, pressure pool operators or force domestic miners to cooperate. That would reduce the need to purchase millions of new machines, but it would replace an industrial problem with an enormous political and coordination problem. Bitcoin mining is spread across several jurisdictions precisely because miners have historically relocated when governments became hostile.

The attack might remain possible. Keeping it invisible would be another matter entirely.

The Short Position May Be More Fragile Than the Attack

Suppose, despite these obstacles, an organization could prepare an attack for $8 billion. It would still need a financial position large enough to recover that investment.

If Bitcoin fell by 50%, a linear short position with $16 billion in notional exposure would produce approximately $8 billion in gross profit. That is before funding costs, trading fees, slippage, infrastructure expenses and potential losses elsewhere.

Leverage would allow the position to be opened with less initial collateral, but it would not make the economic exposure disappear. A leveraged $16 billion short is still a $16 billion position. It is simply more vulnerable to liquidation if Bitcoin rises before the attack begins.

The size of the derivatives market can also be misleading. Open interest measures outstanding contracts, not the amount of money patiently waiting to take the other side of one enormous trade. Every short must be matched by a long, and attempting to build a position representing a significant share of the entire market would alter prices.

Perpetual funding rates could turn sharply negative, making the short increasingly expensive to maintain. Put options would become more costly as sellers detected unusual demand. Futures curves, volatility markets and order books could all reveal that a major participant was preparing for an extreme downside event.

Spreading the position across multiple exchanges, over-the-counter desks and financial products might hide part of the activity, but it would create more counterparties, more collateral requirements and more opportunities for information to leak.

Settlement presents an even greater problem. If Bitcoin truly entered a catastrophic collapse, some exchanges could suspend trading, raise margin requirements or experience counterparty failures. Crypto derivatives platforms rely on liquidation engines and insurance funds to manage bankrupt positions. In extreme conditions, mechanisms such as forced deleveraging may close profitable positions before traders realize their expected gains, as explained in Deribit’s liquidation documentation.

A trade designed to profit from Bitcoin falling to zero therefore contains a paradox: the more successful the attack becomes, the less certain it is that every trading venue will remain liquid, solvent and capable of paying the attacker.

Would Bitcoin Actually Die?

A 51% attack would be one of the most severe crises in Bitcoin’s history. Exchanges would probably suspend deposits, increase confirmation requirements and closely monitor chain reorganizations. Payment processors could stop accepting transactions until the situation became clearer.

But Bitcoin is not a machine that continues operating without human intervention. It is a protocol surrounded by miners, node operators, exchanges, developers, investors and businesses. All of them would have an incentive to protect the value of the network.

Honest miners could add or redirect computing power. Node operators and developers could coordinate emergency software changes. In the most extreme case, the community could consider rejecting the attacker’s chain or changing the mining algorithm, rendering the hostile ASIC fleet useless.

Such measures would be technically and politically controversial. They could split the community and damage Bitcoin’s claim to neutral, predictable rules. Yet they illustrate why an attacker would not be fighting passive code. They would be fighting an ecosystem capable of adapting once the attack became visible.

This does not make Bitcoin invulnerable. A prolonged attack could freeze economic activity and inflict lasting reputational damage. A hostile government might also accept enormous financial losses for political or strategic reasons, making profitability irrelevant.

Harvey’s theory remains valuable because it identifies a genuine blind spot. Derivatives can alter incentives by allowing someone to profit from destruction without extracting that profit directly from the system being destroyed. That principle applies well beyond Bitcoin.

Nevertheless, presenting $8 billion as the price of a Bitcoin kill switch is misleading. That figure does not adequately capture supply-chain inflation, electrical infrastructure, operational secrecy, market impact, collateral, counterparty risk or the network’s response.

To succeed, the attacker would have to dominate semiconductor supply, industrial logistics, energy infrastructure and global derivatives markets almost simultaneously—and do so without provoking an effective reaction.

The real lesson is therefore more nuanced. Bitcoin’s survival is not guaranteed by cryptography alone. It depends on the extraordinary difficulty of coordinating an attack across both the physical and financial worlds.

That may be less comforting than saying Bitcoin is impossible to kill. But it is considerably more realistic.