Newsroom
10 October, 2026 / News / AI / Tags: qureshi, drake, cryptographic, haseeb, bunker

Haseeb Qureshi challenges Ethereum researcher Justin Drake’s bunker mode warning, proposing systemic cryptographic recovery as AI advances raise signature risks for digital assets
Concerns over artificial intelligence accelerating threats to cryptocurrency security intensified this week after an Ethereum researcher flagged potential near-term risks to widely used digital signatures. The discussion has drawn contrasting views from industry figures on how best to prepare.
Justin Drake, a researcher focused on Ethereum protocol development, warned that rapid advances in AI mathematical capabilities could compromise the elliptic curve digital signature algorithm, or ECDSA, which secures many cryptocurrency wallets and transactions. Citing a recent OpenAI report on AI progress in mathematics, Drake stated it is now reasonable to prepare for the possibility that ECDSA could fail before quantum computers achieve a similar breakthrough, potentially in months rather than years in a worst-case scenario.
Drake advised users to gradually shift holdings to new addresses where public keys remain unexposed. Ethereum co-founder Vitalik Buterin agreed the industry should treat risks from AI-accelerated mathematics seriously. He cautioned against rushing to move funds, recommending a measured approach while mitigation options are assessed.
Haseeb Qureshi, managing partner at investment firm Dragonfly, rejected the bunker mode concept as cryptographic doomerism. He argued that relocating assets to fresh addresses offers only limited protection. Coins would stay secure solely while they remain unmoved. In a scenario of widespread exploitation, those holdings could lose value amid market disruption from mass sales of compromised assets.
Qureshi contended that individual address migration fails to address systemic vulnerability. He called instead for blockchain networks to implement proactive defenses rather than rely on reactive user actions.
As an alternative, Qureshi outlined a framework he termed Cryptographic Recovery Mode. Under this approach, users would map hash-based backup signature schemes to their addresses in advance. Validators could then trigger recovery processes if primary cryptographic systems failed, restoring access to affected assets without depending solely on preemptive fund movements.
He stressed that network-level upgrades would deliver broader protection against emerging AI-driven threats compared with isolated wallet adjustments.
Blockchain analytics firm Glassnode reported that more than 31 percent of Bitcoin’s circulating supply sits in addresses considered vulnerable if cryptographic protections were breached. Data showed 6.26 million BTC in such positions. Of that total, about 4.33 million BTC face exposure due to address reuse, a risk that transferring to new addresses could reduce. Another 1.94 million BTC are exposed through specific address formats.
Nearly 1.8 million of the vulnerable BTC are held on cryptocurrency exchanges, accounting for more than half of all balances on major trading platforms, according to Glassnode co-founder Rafael Schultze-Kraft.
| Category | Vulnerable BTC (millions) | Primary Factor |
|---|---|---|
| Address Reuse | 4.33 | Repeated use exposing public keys |
| Address Format | 1.94 | Susceptible address types |
| Held on Exchanges | 1.8 | Platform storage |
No cryptographic break has occurred. The debate centers on prospective risks and differing strategies for strengthening long-term resilience across blockchain networks.









