# This is Justin Drake's 'bunker mode' for protecting cryptocurrencies

> On October 7, Drake called for migrating to fresh addresses to protect cryptocurrencies against the risk of AI breaking ECDSA within months
> By José Maldonado · Cybersecurity · 2026-10-08

Some warnings force you to look at your wallet with different eyes even when you are in a hurry. Justin Drake issued one of those warnings on October 7, 2026, from his personal account on X with a message to the industry. The Ethereum Foundation researcher [called for planning](https://firefly.social/post/x/2107837081313505768)calmly what he called bunker mode, a migration to new addresses whose public key remains hidden behind a hash. His worst-case scenario places an ECDSA break within months rather than years, with key recovery in around a week using a GPU cluster.

The context helps explain why Drake is speaking now and why he mentions artificial intelligence before quantum computing. During the same week, OpenAI published 722 mathematical manuscripts generated by an internal model, a volume Drake interpreted as a sign that mathematical superintelligence is already here. Elliptic curves have more mathematical structure than hash functions and could therefore be more sensitive to a new theoretical breakthrough. Drake clarified that no public attack exists today capable of extracting a private key from a public key on the curves used by Bitcoin and Ethereum.

The practical implication looks to October 2026 and the following months, with a dual message worth keeping in mind from the outset. Drake recommends that large holders move most of their funds to addresses that have never signed a transaction. At the same time, he asks that no one act out of panic or in a hurry, because a rushed migration can create custody errors more serious than the theoretical risk. This balance between calm preparation and operational prudence shapes the entire subsequent debate in Europe.

## Justin Drake's call to protect cryptocurrencies

Justin Drake published his [**warning**](https://x.com/drakefjustin/status/2107837081313505768) on October 7, 2026, from his personal account on X in a calm tone. He asked the industry to start planning a controlled migration of assets to fresh addresses. His definition of a fresh address is simple for any wallet user and does not require switching programs.

![Justin Drake's call on X](https://content.observatorioblockchain.com/wp-content/uploads/2026/10/justin-drake-peticion.webp)
*Justin Drake's call on X*

It is an address whose public key remains hidden behind a hash because it has never signed a transaction. When an Ethereum account sends funds, its public key becomes visible so the network can verify the ECDSA signature over secp256k1. When an account has only received funds, the network only knows the hash and not the full key.

Drake proposes moving most of the funds to that second type of address and repeating the move after each signature. Every time an address signs and leaves a remainder, that remainder should travel to another new address. The process can be done from the same seed phrase if the wallet allows it.

*   Addresses that never signed and keep the key hidden behind the hash
*   Addresses that have already signed and leave the key visible on-chain
*   New test addresses with minimal sends to verify backups

## A far-reaching call

The initial scope targets large holders and experienced operators, not the small user moving modest amounts. Drake called on major custody and trading platforms to review their cold storage. He mentioned Binance first and then cited other firms of similar size with a long operating history.

The idea is for teams with more funds to prepare procedures and test new wallets without external pressure. Retail users can observe and prepare their own list of addresses without moving everything at once. No one needs to buy new hardware or install unknown programs during this inventory phase.

The message includes a clear warning against haste and against improvised changes amid volatility. Drake wrote that a rushed migration would do more harm than good because of the risk of copying a seed incorrectly. His recommendation does not claim that current cryptography is broken or that a proof of concept exists in attackers' hands.

**[Asking AI for more privacy requires knowing how to use Ethereum](https://observatorioblockchain.com/en/ethereum/ia-privacidad-ethereum-zkapi/)**

## Why is AI a concern before quantum computing?

Public debate usually places Q-day as the moment when a quantum computer breaks public-key cryptography. Drake shifts the focus to a different scenario that is closer from a theoretical standpoint. His hypothesis is that a highly capable system could discover a classical algorithm that drastically cheapens key recovery.

That algorithm would run on conventional hardware and would not need logical qubits or quantum error correction. The threat would not come from a physics lab, but from an advance in pure mathematics assisted by machines. That is why Drake speaks of post-AI cryptography and not only post-quantum cryptography.

Elliptic curves contain a rich algebraic structure that allows deep reasoning and potential shortcuts. Hash functions in the SHA or BLAKE families offer fewer footholds for that kind of breakthrough. Drake argued that this difference makes ECDSA and RSA more sensitive targets in the face of the new mathematical output.

The data point that sparked the debate came with the 722 manuscripts published by OpenAI in the first week of October. Each manuscript would have required an average of about three hours of reasoning equivalent to ChatGPT Pro. Many proofs appeared in machine-checkable form with Lean. The figure does not prove immediate danger, but it shows a pace hard to imagine three years ago.

The worst case described by Drake puts concrete numbers on planning without speculating in the abstract. He spoke of months rather than years as the surprise window, and of recovery in around a week with a large GPU cluster. His position is that Ethereum is already heading in the right direction with hashing and formal verification, as reported by [pq.ethereum.org](https://pq.ethereum.org/).

## How much money is exposed today?

Current exposure does not depend on predictions about artificial intelligence, but on keys already visible on-chain. Every time an address signs, its public key is recorded for future review. Addresses that have only received funds keep their key hidden behind the hash. This distinction explains why Drake insists on addresses that have never signed.

Project Eleven maintains a [**public list called Risq**](https://bitcoin-risq-list.projecteleven.com/) that tracks more than 14 million Bitcoin addresses with exposed keys. Drake cited that list in his thread and called for special attention to custodians with a long history of signatures. A report cited this week estimates that some 6.04 million BTC had their key exposed in May. Another estimate raises the figure to 6.9 million due to methodological differences.

Address status

What an observer sees today

Exposure level to an advance

Only received funds and never signed

Only the address hash

Lower, key still hidden

Signed once and holds a balance

Public key visible on-chain

Higher, ready for a future attack

Exchange cold wallet with long history

Many old keys and signatures

High, priority for review

New one created for bunker mode

Only the hash until the first spend

Minimal until the first signature

The Satoshi shield concept helps understand the likely order of targets in an extreme scenario. Drake recalled that there are around 20,000 addresses from the early blocks, each with 50 BTC. A rational attacker would start with those large, concentrated targets before looking for small wallets. Drake called that cushion the Satoshi shield for wallets with less than 50 BTC.

The official Ethereum guide maintains in October 2026 that no quantum computer can break the network today, as explained by [ethereum.org](https://ethereum.org/roadmap/security/quantum-resistance/). That guide coexists with Drake's personal recommendation, which calls for preparing an inventory without changing the entire operation. One marks the current technical state and the other proposes cheap insurance against a surprise.

**[Ethereum shelves burning staking rewards after backlash](https://observatorioblockchain.com/en/blockchain/ethereum-retira-quema-staking-eip8363-hegota/)**

## The reaction of the most critical voices

The most cited response came from the hardware wallet industry and focused on operational risk. Ledger's technical lead warned that a massive, rushed migration can cause costly errors. His argument does not deny Drake's theory, but reminds that custody fails more often due to human carelessness. With users managing multiple chains, that reminder carries weight.

Drake anticipated that objection and therefore limited his call to calm planning rather than immediate migration. He asked to start with large, sophisticated holders, with teams capable of testing new flows in controlled environments. He recommended generating new addresses from the same seed phrase when the model allows it. He also asked to move the remainder after each signature so as not to leave a balance in used addresses.

The warning includes a note for signers who run live services and not just static savings. Drake suggested rotating ECDSA keys with each message signed in oracles and layer-two security councils. As an alternative, he proposed adding multisignature with a hash-based scheme such as SPHINCS. The proposal acknowledges that those signers cannot hide, because their work requires signing daily.

The Europol report published this same week reinforces part of the diagnosis with institutional language. The document concludes that blockchains cannot be attacked today by quantum computing thanks to the strength of the hashes linking blocks. The same report points to wallets as the main point of exposure and recommends preventive migration. For readers in Spain and the European Union, that coincidence elevates the issue to a management matter.

## What could change in Europe if the timetable accelerates

Ethereum's official timetable places central post-quantum infrastructure around 2029, with external reviews from January 2027. Drake stated that those deadlines should be reviewed and accelerated in light of the new automated mathematics. His message does not change the approved forks, but adds pressure to prioritize hashing and formal verification. In Europe, that pressure could translate into more audits and more questions from regulators.

Individual users can turn the warning into a simple routine without technical jargon or new purchases. The first step is to take inventory of addresses that have already signed and of balances in used addresses. The second step is to create a small group of new addresses and test minimal sends to verify backups. The third step is to document the process on paper with double verification of the seed.

Companies with self-custody have a more demanding list with clear owners within the organization. They need to review cold storage and separate roles between whoever proposes and whoever signs. They also need to define written thresholds for moving from planning to partial migration. Drake cited major platforms because a failure there would affect thousands of clients at once.

The debate will remain open for months because it combines mathematics and engineering with different incentives. Drake offers cheap insurance with fresh addresses and calls for calm so as not to turn the insurance into an accident. Ledger reminds that the insurance also has an operational cost and that moving funds without method can prove expensive. Europol provides the institutional framework by pointing to wallets as the weak link in October 2026.

**[Industrial quantum computing is cited in Burgos on October 14 and 15](https://observatorioblockchain.com/en/tech-events/la-computacion-cuantica-industrial-se-cita-en-burgos-el-14-y-15-de-octubre/)**
