Cryptography Doomsday Alert: When AI Sets Its Sights on Your Private Keys, How Can Blockchain Save Itself?

marsbitPublished on 2026-10-10Last updated on 2026-10-10

Abstract

Cryptocalypse Warning: How Blockchain Can Survive AI Targeting Private Keys The rapid advance of AI in mathematics, highlighted by OpenAI's breakthroughs, has sparked fears it could soon target and potentially break foundational cryptography like ECDSA. This poses an existential threat to blockchain, where compromised public-key cryptography would mean anyone could steal funds, with no possibility of recovery like traditional systems. While the probability of such a catastrophic break is low, the risk is non-zero and requires immediate, coordinated preparation at the protocol level, not individual "bunker mode" actions. The article argues that blockchain governance must evolve for a "wartime" footing, prioritizing speed over perfect decentralization to react to potential AI-driven cryptographic breaks. It proposes a "Cryptographic Recovery Mode" for major blockchains. This mode would involve: 1. Implementing a mandatory, simple hash-based signature scheme as a fallback for all addresses. 2. Establishing a pre-defined emergency switch for validators to activate if ECDSA is compromised. 3. Upon activation, the chain would switch to the slow but secure hash-based system. 4. A recovery mechanism using zero-knowledge proofs or time-limited Proof-of-Work challenges would allow users to migrate assets, creating a race against potential attackers. The core message is that the industry must act now to build systemic resilience. While likely nothing will happen, the goal is to...

Author: hosseeb, Partner at Dragonfly

Compiled by: AididiaoJP, Foresight News

Two days ago, OpenAI released 722 mathematical manuscripts in one go, claiming it to be "the most important moment in mathematical history." Foundational problems that had puzzled mathematicians for over a century were solved one after another, with each problem consuming only 3 hours of computing power. Mathematicians were stunned—they could no longer predict what would happen next.

It now appears that cryptography might be the next target.

Scott Aaronson stated that OpenAI has already set its sights on breaking cryptography.

The encryption industry is now facing the "mathematical doomsday" seriously—what happens when super-intelligent AI clusters target foundational cryptography?

Matthew Green responded to @kmad and @matthew_pines on October 8, saying: "I think we might lose public-key cryptography."

Let's be clear: we're not talking about AI accelerating the quantum computing process. That is certainly possible, but quantum computing requires significant physical engineering, which AI can't accomplish overnight. This process takes time and is publicly observable. Preparing for the quantum era is crucial, but the path is relatively clear.

We're talking about some unexpected mathematical breakthrough—something that overturns the foundational assumptions of classical cryptography, allowing ordinary computers to compute discrete logarithms and thus break modern public-key cryptography.

This would be an existential-level disaster.

Justin Drake said on October 7: "Today, I call on the blockchain industry to calmly begin planning a 'bunker mode.' My personal suggestion is to initiate a controlled mass migration of assets to new addresses, specifically addresses where public keys remain hidden behind hashes."

In his now-viral "bunker mode" post, Justin essentially advocated: save yourself, protect your own token assets. The disaster might strike others, but at least you know you're safe.

This is cryptography doomsdayism. It's equivalent to saying, since you know the world is ending, start hoarding supplies.

The problem with doomsdayism isn't its core judgment. Of course, there exists a tail risk of catastrophic failure. You can't simply say the risk is zero.

The problem with doomsdayism lies in its response: if this is true, what should we do?

We cannot know the probability that public-key cryptography has some fundamental weakness. It might be low, but it's not zero. For the sake of discussion, let's assume this possibility is 5%. This means there's a 95% chance everything is fine. We'll blog about it, hesitate, and ultimately nothing bad will happen.

In fact, in that 95% scenario, AI will most likely simply confirm our existing belief: public-key cryptography is secure.

But if I told you now that I have a cryptographic scheme with a 95% chance of protecting your funds, you certainly wouldn't place complete trust in it. Cryptographers measure security risks by failure probabilities as low as 2^-128, meaning "almost impossible to fail," not "probably won't fail."

Some might say: true, but if ECDSA is broken, society faces far bigger problems than blockchain! Think banks, TLS, certificate authorities—blockchain is the least of your worries.

But you shouldn't take that as comfort. Banks can recover, TLS can recover, certificate authorities can recover. They can redo KYC, switch to new cryptographic standards.

The encryption industry cannot. If public-key cryptography fails, blockchain cannot recover. In the worst case, anyone could compute anyone else's keys, meaning anyone could take anyone else's money. We wouldn't know who owns what, and there would be no way to prove anything. It would no longer be cryptocurrency but a graffiti wall, forcing us to start over.

Therefore, we must insure against this failure. We cannot guarantee cryptography won't collapse, but we can guarantee it won't destroy blockchain. This "optimization" must happen both at the protocol level and the societal level, requiring immediate action and coordination.

The common saying is "this is not a drill." But in reality, this is an absolutely necessary drill.

We are like countries that must conscript everyone and teach them to shoot. War might come, and you probably won't be sent to the front lines. But if you are, we need you to become a real soldier immediately.

The opposite of doomsdayism is decisive action. Don't retreat; prepare to withstand potential disaster, which means we must start making changes now.

Why Governance Must Change

Historically, our confidence in cryptography was built on a social process.

Cryptographers would spend years trying to attack a structure. If it wasn't broken, we'd say it's probably fine. Surviving a decade, we'd say it's almost certainly fine and place absolute trust in it for production use. But almost all cryptographic schemes are eventually broken—attack methods improve, security margins shrink, and eventually, we must switch to something better.

This pace has never been fast for blockchain: migrations could take years, and social consensus kept up with cryptographic changes.

But this premise is faltering. We are about to face cryptanalysts stronger than any human, scrutinizing these schemes with a rigor never seen before. There are perhaps only a few thousand truly qualified cryptanalysts in the world, and that number is about to increase by several orders of magnitude.

All the cryptanalytic results accumulated by humanity over the past 40 years might pale in comparison to what AI will do next. Work that used to take a decade might be done in months. Or even overnight.

The pace that blockchain governance has grown accustomed to simply won't work in that environment.

We are now in cryptowar times; peacetime norms no longer apply. We must be able to act extremely fast because things can change extremely fast. If social consensus can no longer keep pace with the speed of technological and cryptographic change, then social consensus must adjust immediately.

But Won't This Cause Panic?

Vitalik said on October 8: "I don't recommend anyone rush to move funds to new wallets today. But we should take seriously the risk that AI-accelerated math poses to cryptography, and minimize our exposure not just to quantum-vulnerable cryptography, but potentially AI-vulnerable cryptography as well."

Panic is not the answer either. I agree with Vitalik: if we encourage panic, people are more likely to cause losses during hasty migrations, which is worse than the actual risk of ECDSA keys being compromised.

So-called "bunker mode" is not a real way out either. You alone moving your keys to a new address might make you safe. But then what? The crux of bunker mode is: your coins can't move. As long as you don't move them, they're protected. But if the entire chain collapses, what use is your individual safety? If everyone's coins are being stolen and dumped, your coins are worthless. Even if your address remains untouched, what's the point?

No, the solution must be systemic. Encouraging people to fend for themselves doesn't solve the problem.

At the same time, the argument that "the most important thing is for everyone not to panic" is becoming less tenable. We should be honest and direct about the risks, get people to take them seriously, and push them to take action.

This keeps reminding me of the Cold War. During the Cold War, Western societies were repeatedly told: nuclear war could happen. For most people, civil defense drills seemed pointless—hiding under a desk wouldn't stop a nuclear bomb. But precisely because everyone participated in the drills, the entire society truly understood the severity of the risk. Political winds shifted toward avoiding war at all costs while building bomb shelters just in case. And indeed, we came close to nuclear war several times. We were truly lucky. The stories of the Cuban Missile Crisis and Stanislav Petrov show: the probability was not zero.

In the matter of cryptography, we might be lucky. Most likely, we will be, but we should be prepared in case luck is not on our side.

So, how must governance change?

1) In a crisis, speed is more important than decentralization. Chains that can coordinate quickly will have an advantage. Decentralized social consensus is inherently mismatched with the pace of ultra-fast technological change. Governance mechanisms must adjust as soon as possible. Even if this means loosening decision-making rules, even if this means giving more power to validators, it must be done. Potential crises require swift and decisive responses.

2) The entire ecosystem must be mobilized. It's not just protocol developers; wallets, exchanges, RPC providers, applications, asset issuers, and end-users all need to act in unison; the entire ecosystem must respond together.

3) Plans must be made in advance. If things fall apart and there's no plan, it will be too late.

The market will watch which chains take this seriously. I've already received inquiries from large investors who understand why this is concerning and want to know if our entire industry has an answer.

So, let's provide an answer.

Cryptographic Recovery Mode

I propose that every major blockchain implement a mechanism as soon as possible, which I call "Cryptographic Recovery Mode." This is a nuclear-level contingency plan to be used only in the worst-case scenario.

The general idea is as follows:

In the next protocol upgrade, add a brand-new, extremely simple hash-based public-key system. (Hash-based signatures typically rely on the fewest cryptographic assumptions.) This is akin to Cold War-level minimalist cryptography. This scheme is slow, expensive, and extremely cumbersome to use. We don't want to touch it at all—almost anything else is better if we have a choice. If we actually have to use it, the chain will be as slow as a snail, and DeFi will be largely paralyzed.

But it works. Technically functional is better than nothing.

Everyone creates a backup key using this terrible signature scheme. All users will be guided to establish a link between their regular address and this hash-based backup key. Initially optional, it becomes mandatory after a few months. Once enforced, you cannot sign new transactions with your ECDSA key until you set up your hash-based backup key. Without a backup key, transactions will simply fail.

Every exchange will also start generating backup keys for their addresses. Coinbase, Metamask, Ledger—if they detect you lack a backup key, they will prompt you to create one before you sign. This way, everyone has a backup plan. Of course, we continue using ECDSA normally.

We pre-set an emergency switch. If disaster strikes and AI discovers a vulnerability in ECDSA, validators can agree to activate the emergency protocol, entering crisis mode. No protocol upgrade is needed; merely by validators signaling, they can vote to flip the switch.

Once the switch is flipped, all addresses enter Cryptographic Recovery Mode. At this point, ECDSA is invalid. Completely dead.

Then What?

At that point, the chain will be a mess. Nothing will run well; it will be almost unusable—but technically, it will still run. Developers will have to figure things out: how much damage is done, which cryptography is still intact, how to transition to a new chain to bring performance back to acceptable levels. At least for now, everyone's balances remain secure, and there is a path to recovery.

What about those who haven't migrated in time? Almost certainly, a significant number of addresses will be left behind.

Those still holding their seed phrase can, at any time, generate a hash-based zero-knowledge proof of their seed phrase (similar to @VitalikButerin's 2024 proposal) and assign themselves a new hash-based address. If their address has never signed any transaction (the so-called "bunker mode"), recovery with a ZK proof is straightforward.

But many simply won't be able to. For them, a countdown begins. To recover, they must take this path: they need to both generate an ECDSA signature and solve a Proof-of-Work hash puzzle to unlock their address and move assets out. The difficulty of this puzzle scales based on the amount of native tokens held by the address. That is, if you hold 100 ETH, the difficulty is 100N.

But this N grows exponentially. 1 day later, it's 100N, 2 days later 200N, then 400N, then 800N. Validators can vote to cap N or limit the growth rate based on the nature of the vulnerability and their risk modeling. Maybe we'll have months; maybe only hours.

To be clear: in a real attack scenario, this becomes a race. If someone actually finds an efficient method to break ECDSA keys, the real owner's only advantage is: the attacker must first crack the key, then do the PoW; the real owner only needs to do the PoW, which should be faster. This makes large-scale theft difficult to execute. If massive theft is detected, validators can drastically accelerate the difficulty growth, preventing more funds from flowing out (or, if things go completely out of control, raise the difficulty to infinity, permanently freezing all vulnerable coins).

I'm certainly missing many crucial details. This is just a proposal; I am neither a cryptographer nor a protocol designer. But the general idea is: a mechanism like this should be implemented on every blockchain as soon as possible.

The market will soon demand this. Given the changes happening in mathematics, every blockchain needs a backup recovery plan, and consensus must be reached in advance—so that when that moment truly arrives, no one is still debating how the transition should happen.

In a crisis, speed is everything.

In such a situation, don't let the perfect be the enemy of the good. If OpenAI is already working on this now, we don't know how soon an unexpected cryptographic breakthrough might occur. The first version of Cryptographic Recovery Mode can be rough (and probably should be rough), but once a preliminary scheme is in place, it can be refined. (Note: if a sudden quantum breakthrough occurs, this mechanism also applies. But its primary need is to address the fundamental breaking of underlying cryptography.)

The key is that we must not just debate and argue on Twitter; we must actually get to work.

Nothing Is Permanent

A few weeks ago, I argued that Zcash should complete its post-quantum migration and then remain unchanged.

I retract that statement now. When the mathematical foundations beneath our feet are as shaky as they are now, nothing is permanent.

Perhaps years from now, we will have more confidence in the foundations of cryptography. But for now, everything is still in flux.

This requires us to react faster, trust protocol teams' judgments more, and face the trade-offs before us more honestly. Ultimately, it's about safeguarding the security foundations of the entire crypto industry amidst rapid technological change.

The "mathematical doomsday" of cryptography is approaching, and we must prepare.

I want to be clear: most likely, nothing will happen, and everything will be fine. AI will invest vast computing power and ultimately conclude—cryptography is all secure enough.

But we must ensure one thing: even in the worst-case scenario, blockchain remains secure.

Related Questions

QWhat is the main threat discussed in the article regarding AI and blockchain?

AThe article discusses the threat of a super-intelligent AI cluster targeting foundational cryptography. It warns of the potential for an unexpected mathematical breakthrough that could allow ordinary computers to compute discrete logarithms, thereby breaking modern public-key cryptography. This is described as an existential-level disaster for blockchain, as it would compromise the security of all keys and assets, rendering the system untrustworthy and unusable.

QWhy does the author argue that the failure of public-key cryptography would be more catastrophic for blockchain than for traditional systems like banks?

AThe author argues that while systems like banks, TLS, and certificate authorities can recover from a failure of public-key cryptography by implementing new standards and re-verifying identities (e.g., through KYC), blockchain cannot. If public-key cryptography fails, anyone could compute anyone else's private keys, meaning anyone could steal anyone's assets. This would destroy the fundamental trust and property rights of the system, making it a 'graffiti wall' where nothing can be proven or owned, necessitating a complete restart from scratch.

QWhat is the 'Cryptography Recovery Mode' proposed by the author?

AThe 'Cryptography Recovery Mode' is a proposed worst-case contingency plan for major blockchains. It involves implementing a new, extremely simple hash-based public-key system in the next protocol upgrade. Users would be required to create and link a backup key using this system to their standard ECDSA keys. In the event of a cryptographic catastrophe, validators could vote to trigger an emergency switch, disabling ECDSA and switching all addresses to this slower, more cumbersome but secure hash-based system. This would allow the chain to remain technically functional, protecting user balances while a transition to a new, performant chain is developed.

QAccording to the article, what changes are needed in blockchain governance to prepare for this potential crisis?

AThe article argues that blockchain governance must adapt to prioritize speed over decentralization during a crisis, as the pace of technological change may outstrip the speed of decentralized social consensus. It calls for mechanisms that allow for rapid, coordinated action, potentially by granting more power to validators. Additionally, the entire ecosystem—including protocols, wallets, exchanges, and users—must be mobilized and aligned with a pre-established plan. Governance must be prepared to act decisively and quickly, with plans formulated in advance rather than during the crisis itself.

QHow does the author differentiate between the threat from AI-driven mathematical breakthroughs and the threat from quantum computing?

AThe author differentiates the threats by their nature and timeline. The threat from AI is an 'unexpected mathematical breakthrough' that could rapidly and secretly undermine classical cryptography, allowing ordinary computers to break systems like ECDSA. This could happen suddenly and without much public warning. In contrast, the development of quantum computing, while potentially accelerated by AI, requires significant physical engineering progress that is public and takes time. Preparing for the quantum era is seen as important but follows a relatively clear and observable path, unlike the unpredictable risk of an AI-driven mathematical discovery.

Related Reads

Tech Veteran: Anthropic IPO Will Be the Peak of the US AI Rally, Will "Go Big on China AI" If Conditions Allow

Veteran tech investor Jack Selby, head of Copper Sky and managing partner of Peter Thiel's family office, warns that Anthropic's planned IPO—aiming to raise up to $100 billion—could mark the peak of the AI market rally. He cautions that if the highly anticipated IPO prices well below expectations, the expiry of lock-up periods could trigger significant selling pressure and a large-scale AI market correction. Selby names SoftBank and Oracle as "canaries in the coal mine," flagging Oracle's higher debt ratio as a bigger risk, and suggests investors needing AI exposure consider Google instead. Selby also questions Anthropic's fundamentals: over 80% of its revenue is non-recurring and vulnerable to competitors closing the technology gap within six to nine months. He notes the business is extremely capital-intensive—requiring costly data, energy, data centers, and chips—and that with interest rates no longer near zero, calling it a trillion-dollar company seems far-fetched. He further highlights Chinese competition, saying DeepSeek and other Chinese AI models trail Anthropic by only about six months, posing a direct threat. Selby says he would "aggressively go long on Chinese AI" if conditions allow, arguing that Anthropic cannot claim all its business must rely on U.S. models, since most AI use cases resemble basic tool applications, leaving room for Chinese alternatives.

marsbit2h ago

Tech Veteran: Anthropic IPO Will Be the Peak of the US AI Rally, Will "Go Big on China AI" If Conditions Allow

marsbit2h ago

Trading

Spot
活动图片