h hoge.gg
Subscribe
BTC$67,432.18+2.34%ETH$3,521.44+1.08%SOL$178.62-0.62%BNB$612.30+0.41%XRP$0.6234-0.18%ADA$0.4521+3.12%DOGE$0.1623+1.86%AVAX$38.71-1.24%LINK$17.84+0.92%HOGE$0.00004120+4.21%
BTC$67,432.18+2.34%ETH$3,521.44+1.08%SOL$178.62-0.62%BNB$612.30+0.41%XRP$0.6234-0.18%ADA$0.4521+3.12%DOGE$0.1623+1.86%AVAX$38.71-1.24%LINK$17.84+0.92%HOGE$0.00004120+4.21%
● Bitcoin & Layer-1s

Taproot Turns Five, and the Quantum Clock Speeds Up

Taproot activated five years ago this week. In the same stretch, three quantum hardware milestones landed, and the upgrade's defining design choice is now its biggest liability.

Five years ago this week, on 14 November 2021 at block 709,632, Bitcoin activated Taproot and then, in the way the network prefers, went quiet. It was the last consensus change the network has managed to ship, and the softest of soft forks: an elegant upgrade that most holders never noticed and still do not think about. The fifth anniversary arrived with exactly that mood.

What made this particular birthday different was the calendar around it. In the same seven days that Taproot turned five, three quantum-computing companies posted new hardware numbers. IBM used its developer conference to unveil a Nighthawk processor and reaffirm a path to a fault-tolerant machine by 2029. A cold-atom startup, Quantum X Labs, published a roadmap aiming at fault tolerance by 2030. And Infleqtion reported 30 working logical qubits, a milestone it said arrived ahead of schedule. None of these machines can touch Bitcoin. All of them are climbing the one curve that eventually meets Taproot’s single most consequential design choice.

Because the thing Taproot did, the thing that made spending more private and more efficient, is also the thing that makes Bitcoin’s quantum question a present-tense problem rather than a someday one. Taproot put public keys in the open. This is the story of why the quiet upgrade spent its fifth birthday being argued about in units of qubits, and what Bitcoin is actually doing about it.

The anniversary nobody threw a party for

Taproot was never one change. It activated as a bundle of three Bitcoin Improvement Proposals: BIP 340 introduced Schnorr signatures, BIP 341 defined the Taproot output type and its Merkle tree of scripts, and BIP 342 added the Tapscript rules for spending them. The authors read like a roster of Bitcoin cryptography (Pieter Wuille, Jonas Nick, Tim Ruffing, Anthony Towns), and the activation itself used a compressed signaling method called Speedy Trial: lock-in came at block 687,284 in June 2021 with more than 90 percent of miners signaling, and the rules went live at block 709,632 that November.

Five years on, it is still the end of the line. No soft fork has activated since, which makes the stretch from late 2021 to late 2026 the longest upgrade drought in Bitcoin’s history. The adoption picture is its own paradox. During the Ordinals and Runes frenzy of early 2024, Taproot spending climbed above 40 percent of transactions; by the autumn of 2026, with speculative inscription demand drained away, Taproot inputs appeared in well under 10 percent of transactions, and P2TR’s share of addresses slid from roughly 54 percent at the 2024 peak into the low 20s. The upgrade is quietly embedded in modern wallets, Lightning channels and token protocols even as its raw on-chain footprint shrank.

What Taproot actually changed

The headline was Schnorr. Where Bitcoin’s original ECDSA signatures do not add together cleanly, Schnorr signatures are linear: two signatures can be combined into one valid signature for the sum of two public keys. That single property lets a group of signers, a two-of-three vault or a Lightning channel, produce one 64-byte signature against one 32-byte public key that looks, on-chain, exactly like an ordinary single-signer spend. Multisignature stopped announcing itself.

The second half was structural. A Taproot output (the kind behind a bc1p address) commits to a public key that is secretly also the root of a Merkle tree of alternative spending conditions. If every participant cooperates, they spend through the key path and reveal nothing but a signature. If they fall back to a scripted condition, they reveal only the one branch they used, never the others. Privacy, efficiency and programmability all improved at once.

A Lightning channel is the cleanest example. Two parties fund a shared two-of-two output; while they cooperate, every update and the eventual close looks like a plain single-signature Taproot spend, indistinguishable from one person moving their own coins, and the dispute scripts that protect each side stay hidden unless someone actually tries to cheat. That is the elegance Taproot sold in 2021: the happy path is private and cheap, and the complexity surfaces only when it has to. The catch, invisible then and loud in 2026, is where that public key sits.

The design choice that put public keys in the open

Here is the crux the anniversary coverage mostly skips. In a Taproot output, the thing written to the blockchain is, in effect, the public key itself (tweaked, but recoverable). It is visible from the moment the output is funded, before it is ever spent. Older address types do not work this way. A legacy pay-to-public-key-hash output and a native SegWit pay-to-witness-public-key-hash output both commit to a hash of the public key; the key only appears when you spend. Until that moment, an observer sees a fingerprint, not the key.

That difference is the whole quantum story. Shor’s algorithm, run on a large enough error-corrected quantum computer, can derive a private key from an elliptic-curve public key. If the public key is hidden behind a hash, an attacker has to break the hash first, and hashing enjoys only a quadratic speedup from Grover’s algorithm, which is nowhere near enough to matter. If the public key is already in the open, the attacker can aim straight at the private key. Taproot, by design, leaves it in the open. As Google’s researchers put it in March, the upgrade made public keys visible on the blockchain by default, removing a layer of protection that older formats kept. It was a rational trade in 2021, when a cryptographically relevant quantum computer was a distant rumor. It is a live question now.

Three quantum milestones in one week

The reason the trade feels live is the news flow that happened to land on top of the anniversary. In mid-November, at its developer conference, IBM presented its latest Nighthawk processor and reaffirmed its published roadmap toward Starling in 2029, a fault-tolerant machine designed to run 100 million gates across 200 logical qubits, with a larger Blue Jay system (2,000 logical qubits) targeted for after 2033. On 12 November, Quantum X Labs put out a cold-atom roadmap aiming at fault tolerance by 2030. And on 16 November, in its quarterly results, Infleqtion reported that it had entangled 30 logical qubits out of 80 physical ones on its Sqale system, crediting AI-assisted design with halving the gates needed for a key operation, and set its next target at 100 logical qubits in 2028.

Chief executive Matt Kinsella called 30 logical qubits an important step on our technology roadmap. That is the honest register for all three announcements: steps, not arrivals. A word of caution before anyone panics or celebrates, though. Logical-qubit counts are not comparable across companies. A logical qubit is an error-corrected unit assembled from many physical ones, and the quality of that correction (the logical error rate) differs enormously between superconducting chips, neutral-atom arrays and trapped-ion machines. Thirty logical qubits on one platform is not thirty on another. What the week shows is direction and pace, not a finish line.

Milestone (Nov 2026)DateHeadline figureWhat it is
IBM Nighthawk / Starling roadmapMid-Nov200 logical qubits by 2029Superconducting; Starling is the first fault-tolerant target
Quantum X Labs roadmap12 NovFault tolerance by 2030Cold-atom platform, roadmap to roughly 800 logical qubits
Infleqtion Sqale16 Nov30 logical qubits (from 80 physical)Neutral-atom; next target 100 logical in 2028
Reference: breaking Bitcoin signaturesGoogle, Mar 2026~1,200 to 1,450 logical qubitsEstimated threshold to run Shor against an exposed key

The gap between 30 and 1,300

Put the numbers side by side and the distance is obvious. Google Quantum AI’s March research estimated that breaking Bitcoin’s signature scheme would take roughly 1,200 to 1,450 high-quality (logical) qubits and fewer than 500,000 physical qubits, and that such a machine could forge a transaction while it sits in the mempool in about nine minutes, giving it roughly a 41 percent chance of beating the original spend into a block. Against that bar, the best public logical-qubit count in November 2026 (about 30, from Infleqtion) is some 40 times short. IBM’s 2029 Starling target of 200 logical qubits is still a factor of six away. On IBM’s own published roadmap, the first machine with enough logical qubits to credibly threaten the signatures is Blue Jay, slated for after 2033.

So why not relax? Three reasons keep the clock ticking. First, roadmaps have been landing early, as Infleqtion’s own ahead-of-schedule framing shows. Second, the requirement itself keeps falling: Google’s sub-500,000-qubit figure was a roughly twenty-fold cut from the earlier conventional estimate of millions, and analysts now describe a countdown in which the estimates keep falling. Third, AI is compressing the engineering, as Infleqtion’s gate reduction hints. The threat is not the hardware that exists; it is the slope of two lines, capability rising and the required threshold dropping, and the point where they cross.

One distinction matters more than any single qubit count. The acute danger is Shor against exposed signature keys, not Grover against hashing. SHA-256, the function behind both mining and hash-based addresses, loses only a quadratic factor to Grover, which leaves it comparatively resilient; that is why a cryptographically relevant quantum computer is a signature problem long before it is a mining problem, and why the hashrate story of 2026, miners squeezed as Bitcoin trades below the cost of production, is a separate worry from this one. Quantum does not out-mine the network first. It out-signs a vulnerable owner first.

Why 6.9 million BTC are already in the open

Taproot did not invent exposed keys, but it normalized them. The total pool of Bitcoin sitting on addresses whose public keys are already visible is large: Google’s analysis put it at about 6.9 million BTC, roughly a third of the supply, including some 1.7 million from the early pay-to-public-key era (coins frequently attributed to Satoshi Nakamoto and other first-year miners) plus everything leaked through address reuse, plus every Taproot output ever created. These coins do not need to be spent to be vulnerable. The key is already on display.

The industry has started counting out loud. In mid-October, nine firms formed a Quantum Alliance and estimated that 6.4 million BTC, about 32 percent of supply and including roughly 1.1 million attributed to Satoshi, sit on exposed public keys, and pegged the earliest credible attack window at around 2028. The exact figure moves with methodology, but the order of magnitude does not: somewhere near a third of all Bitcoin is, in the quantum sense, pre-exposed. Taproot’s contribution was to make funding an output and revealing a key the default behavior of the newest address type, turning a legacy quirk into an ongoing habit.

There is a subtlety worth drawing out. The familiar quantum worry, harvest now and decrypt later, is really an encryption problem; Bitcoin’s exposure is simpler and in some ways worse. An attacker does not need to capture or store anything, because an exposed public key is already permanent public data. They only need to wait for a machine capable of inverting it, then spend. That is why dormant coins on exposed keys, the lost wallets and the earliest mined blocks, are the softest targets: no owner is watching, no owner will move them, and the key has sat in plain view for more than fifteen years.

Two windows: coins in flight and coins at rest

It helps to separate two different attacks, because they threaten different coins. The first is the in-flight window. When you spend any address type, legacy or Taproot, your public key enters the mempool along with your transaction. A fast enough quantum attacker could derive your key and broadcast a competing, higher-fee transaction that steals the funds before yours confirms. That is the nine-minute, 41 percent scenario, and it touches even hash-protected addresses, because at the moment of spending the key is revealed regardless.

The second is the at-rest window, and it is the one Taproot widened. An output whose public key is already visible (a Taproot output, a reused address, an old pay-to-public-key coin) is exposed continuously, with no spend required. A future attacker with a working machine could drain it at leisure, in no particular hurry, picking the fattest targets first. A single-signature bc1p output you funded once and never reused is still permanently exposed, where an unreused hash-based output would only expose its key at spend. The practical mitigations today are old advice made newly urgent: never reuse addresses, and keep long-term cold storage in hash-locked types. Taproot, by construction, cannot offer the hidden-key property to the coins it already holds.

For most holders in 2026 this stays theoretical, and the sensible response is proportionate rather than panicked. Spreading coins across address types, avoiding reuse, and keeping the bulk of a stack in unreused hash-based cold storage all shrink the at-rest exposure without betting on any particular timeline. Where you keep keys matters too, because a quantum migration will eventually ask every wallet to generate and back up a new kind of address, so the wallet you standardize on today is quietly also a choice about how painlessly you migrate later. None of this is urgent now. All of it gets easier if you start tidy.

Bitcoin’s answer: BIP-360 and BIP-361

The protocol-level response has two moving parts. The first is an address framework. BIP-360, authored by Hunter Beast with Ethan Heilman and Isabel Foxen Duke and merged into the BIPs repository in February 2026, defines a new output type nicknamed Pay to Merkle Root (P2MR). It does not itself add a post-quantum signature; it builds the address structure that hides the signature scheme behind a Merkle commitment, so a hash-based or lattice-based scheme can be slotted in later. Like Taproot, it would activate as a soft fork, and old software would keep working.

One thing BIP-360 deliberately does not do is pick the post-quantum signature scheme. That choice is contested in its own right: hash-based schemes such as SPHINCS+ are conservative and well understood but produce very large signatures, while lattice-based schemes are more compact and rest on newer assumptions. By hiding the scheme behind a Merkle commitment, the address framework lets Bitcoin defer the decision, and even support more than one option, which matters because whatever it picks will be measured in kilobytes rather than Schnorr’s 64 bytes and will set the cost of every post-quantum spend for years.

The second part is migration, and it is harder. BIP-361, a Post-Quantum Migration and Legacy Signature Sunset proposal from Jameson Lopp and co-authors submitted in April 2026, lays out a staged plan: first deprecate sends to vulnerable address types, then, years later, stop honoring old ECDSA and Schnorr signatures from those vulnerable outputs. That second step is the controversial one, because it effectively freezes any coins that did not migrate in time, including coins whose owners are gone. BIP-360 co-author Heilman has estimated the full migration could take about seven years, which is precisely why the argument is happening five years into Taproot and not after the first quantum computer boots up.

Address typeWhen is the public key exposed?Quantum posture at rest
P2PK (earliest coins)Always, from creationWorst; key visible for over a decade
P2PKH / P2WPKH, reusedAfter the first spendExposed once reused
P2PKH / P2WPKH, never reusedOnly at the moment of spendingHash-protected while unspent
P2TR (bc1p, Taproot)From the moment it is fundedExposed at rest by design
P2MR (BIP-360, future)Uses a post-quantum signatureResistant, once activated

Freeze, migrate, or do nothing

The cryptography is the easy part. The politics are not, and the Quantum Alliance made the disagreement public rather than hiding it. Jameson Lopp, the BIP-361 co-author and Casa co-founder, has framed the dilemma more cleanly than anyone: in his essay on the migration debate he argues that every serious option violates a principle, setting the old maxim not your keys, not your coins against a new one, your keys, only your coins. A sunset that freezes un-migrated outputs looks like a confiscation to one camp; leaving millions of exposed BTC for an attacker to harvest looks like negligence to the other. There is no option that keeps everyone’s principles intact.

The voices inside the Alliance split along that seam. Christopher Smith, chief executive of Quantus, put the threat in blunt terms, warning that whoever gets the first CRQC will take them, probably with government backing, where CRQC is the cryptographically relevant quantum computer the whole debate orbits. His colleague Bryn Bennett, chief risk officer at The Vault, pushed the decision back onto the network itself: freezing coins, he said, is a governance decision for the Bitcoin network, not for the industry. Adam Back, the Blockstream chief executive, has argued for optional upgrades over any forced freeze and for calm sequencing, telling a Paris conference in April that preparation is key, and making changes in a controlled way is far safer than reacting in a crisis. Changpeng Zhao, for his part, has suggested simply freezing Satoshi’s roughly 1.1 million coins. No date is set, and no consensus exists.

Can Bitcoin even ship a fork? The BIP-54 canary

None of the quantum plans matter if Bitcoin cannot activate a soft fork, and it has not since Taproot. That is the uncomfortable prerequisite sitting under the whole discussion. The nearest test is BIP-54, the so-called Great Consensus Cleanup, a feature-free bug-fix fork that nearly everyone agrees is harmless. Developer Matt Corallo recently put it at about 80 percent to be Bitcoin’s next soft fork, and framed it as a practice run to rebuild the network’s activation muscles. If Bitcoin can rehearse a fork on something uncontroversial, the far harder quantum migration inherits a template. If it cannot, BIP-360 and BIP-361 are stuck no matter how good the cryptography is.

The deeper problem is that Bitcoin never settled on how to activate anything after the 2017 SegWit fight. Speedy Trial got Taproot over the line because the change was uncontroversial, but there is no agreed playbook for a contested one, and a signature sunset that could freeze a third of the supply is about as contested as proposals come. That is why developers treat BIP-54 as a rehearsal: not because the bug fixes are thrilling, but because the network needs to relearn how to agree before it meets a change nobody can opt out of.

The contrast with the covenant proposals is instructive. CTV and CSFS, the opcodes that would unlock vaults and more expressive scripts, have sat at essentially zero percent miner signaling, with the CTV activation window running to 30 March 2027, and OP_CAT remains spec-complete with no activation path at all. The upgrade queue is long, and every item on it competes for the same scarce resource: the social consensus to change Bitcoin’s rules.

ProposalWhat it would doStatus, late 2026
BIP-54 (Consensus Cleanup)Fix four longstanding bugs~80% odds to be next fork; on signet
BIP-119 CTV / BIP-348 CSFSCovenants, vaultsRoughly 0% signaling; CTV window to Mar 2027
BIP-347 OP_CATRecursive, stateful scriptsSpec complete; no activation path
BIP-360 P2MRPost-quantum address frameworkMerged Feb 2026; not activated
BIP-361 sunsetPost-quantum migration and freezeSubmitted Apr 2026; contested

The migration is the risk, not just the machine

Even with a cryptographically relevant machine years off, the migration itself is the dangerous stretch. Three problems stack up. The first is coordination: moving something like 6.9 million BTC spread across millions of outputs, an unknown share of them controlled by keys that are lost and therefore can never be moved at all. The second is size. Post-quantum signatures are far larger than Schnorr’s compact 64 bytes, which revives Bitcoin’s oldest fight over block space and fees; bigger witnesses mean fuller blocks and higher costs exactly when everyone needs to transact.

The third problem is human, and it is the one that will hurt first. A multi-year, network-wide campaign that tells every holder to move funds to a new address type is a gift to attackers. Fake migration tools, spoofed wallet prompts and look-alike addresses are the drainer’s natural habitat, and a quantum migration window is a years-long open season. Anyone planning to move coins should treat the whole period as a heightened-threat environment and lean on a hardened phishing-defense stack rather than trusting any inbound instruction. There is also the quieter adoption drag: support for the new output type will roll out unevenly across wallets, just as bc1p support did, and a key migration moves only as fast as the slowest widely used wallet.

What the price is not saying

The spot market is ignoring all of this, for now. Bitcoin traded around 64,600 dollars on 17 November, down about 0.4 percent on the day and roughly 14 percent on the week, leaving it close to 49 percent below the record near 126,080 dollars set in October 2025, with a market capitalization around 1.3 trillion dollars, according to CoinGecko. The move that mattered was macro and flows, not qubits, and it fits the broader year-end picture of a market staring at its first back-to-back annual loss.

Not everyone thinks the price should ignore quantum. Charles Edwards, who runs Capriole Investments, has floated what he calls a Quantum Discount Factor, arguing that fair value should carry a haircut until the risk is resolved at the protocol level. The on-chain analyst Willy Woo takes the opposite view, telling holders that quantum will not break BTC because BTC will adapt. The honest read sits between them: quantum is a tail risk the spot price does not yet carry, and it is better understood as a governance deadline than a trading catalyst, right up until the moment it is both.

Why no regulator can fix this

It is worth saying plainly, because the instinct in a US market is to ask who is in charge. The SEC regulates securities, trading venues and disclosures; it does not set Bitcoin’s consensus rules, and no regulator approves or blocks a soft fork. BIP-360 and BIP-361 live or die by miners and node operators, not by a filing, which is a different world from the deadline-driven rulebook fights playing out in Washington. The quantum question is one of the few big crypto stories where the agencies are spectators.

For an individual holder, the mechanics are simpler than the governance. Moving your own coins from a bc1p output to a future quantum-resistant address is a transfer between your own addresses, which under current US guidance is generally not a taxable disposal, though you still carry your cost basis across; that is a planning detail worth folding into any year-end tax review, especially in a down market where harvesting decisions interact with wallet hygiene. The protocol will not protect coins their owners refuse to move; the sunset debate exists precisely because some owners never will.

What to watch between here and the first CRQC

The next few years are a race between two clocks: how fast the hardware climbs, and how fast Bitcoin can agree to change. These are the signals that will tell you which clock is winning.

  • Fork machinery: whether BIP-54 activates at all, the first real proof that Bitcoin can still change its consensus rules after five years of drought.
  • BIP-360 progress: a released activation client, signet signaling, or a published timeline for the post-quantum address type.
  • Logical-qubit counts: whether vendors keep beating their roadmaps, with Infleqtion targeting 100 logical qubits in 2028 and IBM aiming for 200 in 2029.
  • The threshold curve: whether academic estimates keep sliding below the current bar of roughly 1,200 logical qubits.
  • Custody posture: concrete migration plans from large exchanges and custodians, and the standards work coming out of the Quantum Alliance.
  • Early-warning tripwires: proposals such as a quantum canary, a deliberately exposed bounty output that a capable machine would drain first, flagging the threat before it reaches real coins.

Five years ago Taproot shipped a cleaner, more private way to lock up Bitcoin. Its fifth birthday is a reminder that the same elegance left the keys on the table. The machines that could read them are still small and still years away, but they are arriving on schedule or faster, and Bitcoin’s answer depends on a fork it has not yet proven it can pass. The quiet upgrade turned out to have set the loudest alarm.

Frequently Asked Questions

Did Taproot make Bitcoin vulnerable to quantum computers?

Not exactly; it changed where the exposure sits. Taproot outputs reveal the public key on the blockchain from the moment they are funded, so those coins are exposed at rest, unlike unreused legacy or SegWit addresses that keep the key hidden behind a hash until it is spent. No quantum computer can break any of them today.

How many qubits would it take to break Bitcoin?

Google Quantum AI estimated roughly 1,200 to 1,450 high-quality logical qubits and fewer than 500,000 physical qubits. The best public logical-qubit count in November 2026 was about 30, from Infleqtion, so the gap is still wide, but the estimates have been falling and roadmaps have been landing early.

Is my Bitcoin safe in a bc1p Taproot address?

For now, yes, because no machine can attack it. But a bc1p output exposes its public key permanently, so for long-term cold storage some holders prefer unreused hash-based addresses and are watching the BIP-360 post-quantum address framework before deciding where to park coins for the next decade.

What are BIP-360 and BIP-361?

BIP-360 adds a post-quantum address framework called P2MR, so new signature schemes can be plugged into Bitcoin later. BIP-361 is a staged migration that would first discourage sends to vulnerable addresses and eventually stop honoring old signatures from them. Both require a soft fork, and neither has activated.

When will quantum computers actually threaten Bitcoin?

Nobody knows. Published hardware roadmaps point to the early-to-mid 2030s for machines large enough, the Quantum Alliance cites an earliest credible window around 2028, and the migration itself could take about seven years, which is why the debate is already underway.

By Marcus Okafor, HOGE Wire senior editor covering Bitcoin’s base layer and Layer 2 ecosystem.

Share 𝕏 Post Telegram