Welcome back! This is Canaan's weekly newsletter on Bitcoin mining, energy, and compute infrastructure.
 

Earlier this month, researchers cut the estimated cost of a key step in a quantum attack on Bitcoin by more than half against Google's March benchmark. The headlines wrote themselves. The new circuit needs 1,151 logical qubits for one calculation inside Shor's algorithm, and it leaves out error correction and the rest of the attack. No existing machine can run it.
 

For scale, the best 2026 hardware demonstrates somewhere between a few dozen and roughly a hundred logical qubits. Breaking Bitcoin's curve needs well over a thousand, running reliably for the full computation. So, while the math is getting cheaper, the machine still doesn't exist.
 

That gap is why serious people land far apart on timing. Blockstream's Adam Back puts a real threat decades out, noting the biggest calculation quantum hardware has managed is factoring 21.

https://x.com/adam3us/status/2039707382960230473

The kind of information that is exposed matters here. The most immediately exposed coins are those whose public keys are already visible on-chain. Google's March paper counted about 6.9 million BTC in that category. CoinShares argues the slice large enough to move markets if stolen is closer to 10,200 BTC, sitting in early P2PK outputs.
 

What quantum does not threaten in any practical sense is proof-of-work. A March study estimated that quantum mining at Bitcoin's January 2025 difficulty would need on the order of 10²³ qubits drawing 10²⁵ watts, against roughly 15 GW for the entire network. The SHA-256 hashing that ASICs like Canaan's Avalon line perform is the part of Bitcoin quantum computing has the least to say about.
 

On yesterday's Canaan America X Space, Danno Ferrin, a core developer at Tectonic Labs focused on post-quantum security, shed some light on the topic, emphasizing that there is time to migrate to post-quantum cryptography (PQC), but we should also get things moving rather than waiting. Danno also shared some concerns about the irreversible exposure of abandoned coins.
 

The defenses are no longer theoretical either. On August 26, StarkWare researcher Avihu Levy's Quantum Safe Bitcoin (QSB) construction was confirmed in block 964,199 with no change to consensus rules. It is a stopgap: each spend costs roughly $75 to $150 of GPU work and only works with older pre-SegWit outputs. Because the format is nonstandard, it had to go straight to a mining pool rather than through the public mempool.

https://x.com/avihu28/status/2092742315995480266

he structural fix is a soft fork. BIP-360, merged into Bitcoin's proposal repository in February, adds a Pay-to-Merkle-Root output that keeps public keys off-chain until spent, with migration left voluntary. BIP-361 goes further, proposing a phased sunset of legacy signatures that could eventually freeze coins that never move. Critics call that confiscatory, and that is where the real fight is.
 

That is the timeline worth watching. Taproot, Bitcoin's last major soft fork, took nearly four years from first proposal to activation in 2021. The quantum question will be settled less by qubit counts than by whether Bitcoin's upgrade process can move ahead of the math, and how quickly the ecosystem, miners included, lines up behind it.

In the News

• Texas Governor Orders Data Center Permit Halt Pending Grid Review

• Fluidstack Starts $4 Billion Texas AI Campus Phase Planned for 1.5 GW

• Virginia Curbs State Support for Large Data Centers, Plans Tighter Rules

• Data Center Power-Cost Bill Stalls in Senate Over Enforcement Dispute

• Nscale's IPO Filing Reveals $103B in AI Contracts with Heavy Funding Needs

• Oracle-Linked $18 Billion AI Data Center Loans Come Under Pressure, FT Reports

• Bitcoin price recovery unlikely to lure AI-focused miners back, CoinShares says

• AI Data Center Developer LIAN Eyes IPO to Raise $500 Million

Network at a Glance

• BTC price (USD): ~$83,972

• Network hashrate: 927 EH/s

• Difficulty: 132.7T

• Hashprice: ~$40.3 / PH / day

Project Spotlight

Three years ago this month, Canaan marked its tenth anniversary in Singapore by launching the A14 series.
 

The venue was Avalon Bitcoin & Crypto Day on September 12, 2023, and the flagship air-cooled model was the A1466: 150 TH/s at 21.5 J/T, drawing about 3,225 W.
 

The context matters. The A1366 before it ran at roughly 25 J/T. Taking another 14% off the energy per terahash in one generation, in a standard air-cooled chassis, meant operators could upgrade without rebuilding their sites around liquid cooling. For fleets working with existing power contracts rather than new infrastructure budgets, that is usually the upgrade that gets approved.
 

The launch also showed where the series was heading. Alongside the A1466, Canaan introduced the A1466I immersion variant at 170 TH/s and 19.5 J/T, taking the series below 20 J/T.
 

A decade in, the A1466 was built to compete on the one metric every operator checks first: joules per terahash.

Events and Media

• Canaan America hosted Danno Ferrin of Tectonic Labs for an X Space, Quantum and Bitcoin: What's Real, What's Hype

• Bitcoin.com News interviewed Gwyn Lauber at Canaan’s site in West Texas

Follow and Contact Us

• X: @canaanio

• Youtube: Canaanmining

• LinkedIn: Canaan Inc.

• Website: canaan.io

• Support: Customer Care
  
  Disclaimer: This newsletter shares industry commentary and third-party news for informational purposes only. The views and opinions expressed by third-party sources are those of their respective authors and do not necessarily reflect the views of Canaan Inc.  For official news, please refer to Canaan’s press releases and SEC filings at https://investor.canaan-creative.com/.