Just saw Google's Quantum AI team drop some pretty sobering research on what quantum computing could actually do to Bitcoin and Ethereum. And honestly, the numbers are way more concerning than what most people have been saying.

So here's the thing: everyone's been throwing around estimates that you'd need millions of qubits to break crypto security. Google's researchers just published findings suggesting it could actually take fewer than 500,000 physical qubits. Even more specific, they've designed attack methods that would only need around 1,200 to 1,450 high-quality qubits. That's a massive gap from what we thought before.

What got my attention is the practical attack scenario they outlined. If someone had a quantum system ready, they could potentially hijack Bitcoin transactions in real time. Here's how it works: when you send Bitcoin, your public key gets exposed for a brief moment. A quantum computer could theoretically use that to calculate your private key and redirect the funds. Under their model, the whole attack could happen in about nine minutes. Bitcoin confirmations typically take around 10 minutes, which means an attacker would have roughly a 41% success rate. That's not theoretical anymore, that's actually concerning.

The research also flags something interesting about Bitcoin's Taproot upgrade from 2021. While Taproot made transactions more private and efficient, it also made public keys visible on the blockchain by default. That actually expanded the attack surface. Google's estimate is that about 6.9 million Bitcoin, roughly one-third of total supply, are already sitting in wallets with exposed public keys. That's a huge number when you think about it.

Ethernet seems less exposed to this specific risk since transactions confirm faster, leaving less time for an attack window. But the overall message from Google is pretty clear: the quantum threat timeline might be shorter than we thought, and the risks are broader.

What's interesting is how Google is handling the security research itself. Instead of publishing step-by-step attack details, they used zero-knowledge proofs to verify their findings without exposing the actual methods. Smart move for preventing misuse while still proving the research is legit.

Google has previously flagged 2029 as a potential milestone for functional quantum systems, so this research suggesting attacks might require less computing power than expected definitely raises the urgency around post-quantum migration. The crypto industry's probably going to need to move faster on this than current timelines suggest. Worth keeping an eye on how projects respond to these findings, especially on platforms like Gate where you can track the market reaction.