Quantum Bitcoin Countdown

Tadge Dryja proposes quantum recovery protocols like 'Lifeboat' to salvage 90% of Bitcoin via soft forks and commit-reveal schemes.

56 min Listen to the original Episode page
Quantum ComputingBitcoin SecuritySoft ForksCryptographyBitcoin Development

Summary

Tadge Dryja and other speakers discuss emergency recovery protocols for Bitcoin in the event of a quantum computing breakthrough. They argue that while proactive migration to quantum-resistant signatures is ideal, 'last-ditch' mechanisms are necessary for dormant or non-migrating coins. Dryja proposes using commit-reveal schemes and soft forks activated by a 'tripwire' transaction to protect assets without destroying them. The discussion highlights the philosophical and technical trade-offs between preventing theft and preserving user sovereignty, particularly for early-era coins lacking seed phrases.

Episode sections

  1. 0:00 – 1:24

    Worst-case scenario and the necessity of recovery protocols

    Speaker_0 argues that a sudden quantum break would cause massive theft but aims for a 90%+ salvageable outcome. They compare recovery protocols to seatbelts: even if people drive better (migrate to secure keys), safety nets are essential for those who don't prepare.

  2. 1:24 – 5:35

    Scope of work: Last-ditch recovery for non-migrating coins

    Tadge Dryja focuses on 'postponed recovery protocols' for coins that failed to migrate to quantum-resistant schemes like Schnorr or Taproot v2. He emphasizes this is a 'last ditch' effort for users who did not plan ahead, such as long-term holders who buried seed phrases.

  3. 5:35 – 10:05

    Philosophical debate: Obligation to protect dormant assets

    Speaker_3 introduces the 'emergency contraception' analogy, contrasting proactive migration with emergency recovery. They cite Yehuda Lindell's view that the community has an obligation to protect assets of users who choose to 'disappear' or stop paying attention to Bitcoin.

  4. 10:05 – 15:27

    Two quantum-secure 'escape routes': Hidden Public Keys and Seed Phrases

    Dryja identifies two secrets a quantum computer cannot easily recover: a hidden public key (not yet revealed on-chain) and a seed phrase (BIP32/BIP39). These rely on hash functions, which remain resistant to quantum pre-image attacks despite Grover's algorithm.

  5. 15:27 – 19:40

    Commit-reveal schemes vs. Zero-Knowledge Proofs

    Dryja discusses 'Dropkick' and his 'Lifeboat' proposal, which use commit-reveal schemes. He prefers these over Zero-Knowledge Proofs (ZKPs) because ZKPs are computationally heavy and introduce new cryptographic assumptions, whereas commit-reveal uses simple, existing hash-based logic.

  6. 19:40 – 24:20

    Implementation via Soft Fork with a Tripwire

    The recovery mechanism is proposed as a soft fork, potentially activated by a 'tripwire' transaction that proves a quantum computer exists. This would impose new signing requirements (like an OP_RETURN proof) on transactions spending from legacy elliptic curve outputs.

  7. 24:20 – 28:52

    Distinction between Public Key and Seed Phrase recovery risks

    Dryja argues that requiring commit-reveal for public keys is safe because the key is already known. However, seed phrase recovery is riskier because many old coins (pre-2013) or imported keys lack seed phrases, potentially rendering those specific coins unrecoverable.

  8. 28:52 – 29:28

    The 90% Salvageability Threshold

    The discussion concludes with a philosophical trade-off: whether it is worth implementing a complex recovery protocol to save 90-95% of coins if it means the remaining 5% (those without recoverable secrets) are permanently lost or stolen.

  9. 29:28 – 31:05

    Philosophical trade-offs and immediate quantum breach scenarios

    Speaker_0 estimates that about 1% of Bitcoin might be destroyed in a quantum breach, mostly abandoned coins with lost keys. Speaker_3 highlights the philosophical challenge to Bitcoin's individualistic ethos regarding responsibility. Speaker_0 describes an immediate breach as a 'huge mess' requiring rapid, potentially unstable software fixes.

  10. 31:05 – 35:12

    Recovery mechanisms and the role of miners

    Speaker_0 explains that without pre-activated quantum-secure signatures, recovery relies on 'ugly' out-of-band methods like embedding commitments in miner Coinbase transactions. He notes that miners on a chain with no recovery plan might simply steal 3 million coins to build the longest chain, creating a chaotic alternative to a chain with recovery software.

  11. 35:12 – 38:24

    Hodler behavior and address reuse

    Speaker_1 asks if post-quantum reality changes how users hold Bitcoin. Speaker_0 argues that users who do not reuse addresses and possess seed phrases are protected because their public keys are hidden. He suggests that proactive migration to new address types via soft forks is possible but not strictly necessary for these users.

  12. 38:24 – 43:12

    Theft versus destruction trade-off

    Speaker_0 argues that allowing some theft is preferable to destroying coins via software restrictions, as developers are less liable for theft than for destroying user assets. He suggests a phased approach where seed phrases are required first to prevent theft, followed by public key revelation later, accepting the risk of theft for those who shared keys.

  13. 43:12 – 45:47

    2009-era coins and 'landmines'

    Speaker_0 states that 2009-era Pay-to-PubKey coins likely have no recovery mechanism because seed phrases did not exist. He discusses Corey Fields' 'landmines' idea, where a bounty is posted to incentivize the original owner to reveal a commitment to spend these coins, though he doubts its practical effectiveness for abandoned coins.

  14. 45:47 – 55:48

    Skepticism on quantum timelines and funding dynamics

    Speaker_0 expresses skepticism about near-term quantum computing, citing the immense engineering difficulty and potential for research funding to be driven by hype rather than physics. He suggests that widespread adoption of post-quantum signatures might reduce the incentive to build quantum computers, potentially delaying their arrival.

Assets discussed

AssetKindStanceContext
BitcoinBTCcryptoneutraleverybody, it's the Quantum Bitcoin countdown today. Saving quantum vulnerable coins.

“Stance” is how the speakers talked about the asset, not a recommendation.

Claims worth verifying

A quantum break would result in 'much higher losses' and 'much more theft,' but the goal is to make it 'sub-livable' by saving ~90% of coins.

Unidentified speaker (speaker_0) ·0:00

How to verify: Primary Source: Transcript [0:00]. Verification: N/A (Speaker's opinion/forecast). For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

There are over six million Bitcoin coins at risk from quantum computing.

Unidentified speaker (speaker_1) ·0:36

How to verify: Primary Source: Transcript [0:36]. Verification: Check on-chain data for UTXOs in legacy address types (P2PKH, P2SH) that have not been spent since the activation of SegWit or Taproot, or estimate based on known dormant UTXO counts. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Tadge Dryja is the co-inventor of the Lightning Network.

Unidentified speaker (speaker_1) ·0:55

How to verify: Primary Source: Transcript [0:55]. Verification: Check the Lightning Network whitepaper (2026) for authorship; Dryja is listed as a co-author. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Tadge Dryja is the creator of Utreexo.

Unidentified speaker (speaker_1) ·0:55

How to verify: Primary Source: Transcript [0:55]. Verification: Check Dryja's GitHub or the Utreexo specification paper for authorship. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Schnorr signatures are a hash-based signature scheme considered quantum-resistant.

Unidentified speaker (speaker_0) ·2:44

How to verify: Primary Source: Transcript [2:44]. Verification: Check cryptographic literature on Schnorr signatures; they rely on the discrete logarithm problem, which is broken by Shor's algorithm, but the speaker claims they are 'hash-based' and 'quantum secure' in the context of the discussion (Note: This is a technical claim by the speaker; standard Schnorr is NOT quantum resistant against Shor's algorithm, but the speaker may be referring to hash-based signatures like SPHINCS+ or mischaracterizing Schnorr. Verify the speaker's specific technical assertion against standard crypto knowledge). For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

There are still transactions that do not use SegWit, despite it being a decade old.

Unidentified speaker (speaker_0) ·4:23

How to verify: Primary Source: Transcript [4:23]. Verification: Check blockchain explorers (e.g., Blockstream, Mempool.space) for the percentage of SegWit vs. Legacy transactions in recent blocks. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

A quantum computer can derive a private key from a public key but cannot reverse a hash to find a seed phrase.

Unidentified speaker (speaker_0) ·11:45

How to verify: Primary Source: Transcript [11:45]. Verification: Cryptographic fact-checking. Shor's algorithm breaks elliptic curve discrete logs. Grover's algorithm provides a quadratic speedup for pre-image attacks but does not break SHA-256/SHA-512 security in the short term. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Grover's algorithm is 'impractical' and 'not parallelizable' for breaking hash pre-images.

Unidentified speaker (speaker_0) ·13:44

How to verify: Primary Source: Transcript [13:44]. Verification: Check recent academic papers cited by Dryja or general quantum computing literature on Grover's algorithm limits. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Bitcoin Core did not support seed phrases until around 2013.

Unidentified speaker (UNKNOWN) ·26:37

How to verify: Primary Source: Transcript [26:37]. Verification: Check Bitcoin Core release notes or BIP32/BIP39 implementation history. BIP32 was 2026, BIP39 2026. Core added BIP39 support later. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

There are coins from 2013-2014 that do not have a seed phrase or derivation path.

Unidentified speaker (speaker_0) ·27:23

How to verify: Primary Source: Transcript [27:23]. Verification: On-chain analysis of UTXOs from that era; early wallets often used raw private keys without HD derivation. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Approximately 1% of Bitcoin coins would be destroyed in a quantum breach, mostly abandoned ones with lost keys.

Unidentified speaker (speaker_0) ·29:28

How to verify: On-chain analysis of unspent outputs from early Bitcoin eras; community estimates of lost coins. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Miners on a chain with no recovery plan could simply take 3 million coins to build the longest chain.

Unidentified speaker (speaker_0) ·38:24

How to verify: Theoretical analysis of miner incentives in a quantum breach scenario; comparison of chain lengths in competing forks. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

2009-era Pay-to-PubKey coins have no recovery mechanism because seed phrases did not exist.

Unidentified speaker (speaker_0) ·43:12

How to verify: Historical review of Bitcoin protocol versions (BIP32/BIP39 did not exist in 2026); analysis of 2026-era transaction outputs. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Speaker_0 is skeptical that quantum computers will arrive in 3 years, citing the need for 'bigger fridges' and billions in funding.

Unidentified speaker (speaker_0) ·46:37

How to verify: Review of quantum computing industry roadmaps and funding reports; comparison of current qubit counts and error rates against requirements for Shor's algorithm. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Bitcoin Research Week in Amsterdam featured significant discussion on post-quantum topics, with some sessions moving away from it due to volume.

Unidentified speaker (speaker_0) ·53:10

How to verify: Conference agenda and proceedings from Bitcoin Research Week Amsterdam; attendee reports or recordings. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Speaker_0's proposals 'Lifeboat,' 'Life Jacket,' and 'Shrink Ray' are available on GitHub at github.com/eabat.

Unidentified speaker (speaker_1) ·54:24

How to verify: Direct visit to github.com/eabat to view repository contents and commit history. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Corey Fields proposed the 'landmines' idea for recovering 2009-era coins.

Unidentified speaker (speaker_0) ·43:44

How to verify: Search for Corey Fields' publications or GitHub repositories related to Bitcoin quantum recovery. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Post-quantum signatures are currently larger (few hundred bytes) than elliptic curve signatures.

Unidentified speaker (speaker_0) ·49:28

How to verify: Comparison of signature sizes in cryptographic libraries (e.g., SPHINCS+ vs. ECDSA); Bitcoin protocol specifications. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Speaker_0 believes that widespread adoption of post-quantum signatures might reduce funding for quantum computing research.

Unidentified speaker (speaker_0) ·50:28

How to verify: Theoretical analysis of incentive structures in quantum computing R&D; review of funding trends in cryptography. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Speaker_0 plans to post his proposals on Delving Bitcoin and Twitter.

Unidentified speaker (speaker_0) ·54:45

How to verify: Monitoring Delving Bitcoin blog and Speaker_0's Twitter/X account for posts. For recent events, verify 2026-08-18 through 2026-11-18. Use another period only when the speaker explicitly names it.

Do your own research

  • Verify the technical feasibility of Grover's algorithm against BIP32/seed phrase security by reviewing recent cryptographic research papers.
  • Check the GitHub repository github.com/eabat for the specific implementation details of the 'Lifeboat' and 'Life Jacket' proposals.
  • Review Bitcoin Core mailing lists and recent conference proceedings (e.g., Bitcoin Research Week) for developer sentiment on post-quantum signature adoption.
  • Analyze the economic incentives for miners in a 'tripwire' scenario to understand potential chain splits or theft dynamics.
  • Research the historical prevalence of Pay-to-PubKey outputs in the 2009-era Bitcoin blockchain to assess the volume of unrecoverable coins.
  • Compare the computational overhead of commit-reveal schemes versus Zero-Knowledge Proofs in the context of Bitcoin block size limits.
  • Investigate the legal and liability implications of software-enforced asset recovery versus theft in various jurisdictions.
Full transcriptexpand

0:00Unidentified speaker (speaker_0): It would be a huge mess if it happened tomorrow. Um so it would be much higher losses, much more theft, but hopefully somewhat sublageable. Everyone can just steal everything. Like, why would a miner even look at a member? They're like, I have like three million points I can just take. Maybe we shouldn't have seat belts because people will be better drivers. They know and they're like, instant death if you crash.

0:21Unidentified speaker (UNKNOWN): Like, But

0:22Unidentified speaker (speaker_0): yeah, okay, but you should still make seed seat belts in airfacts, air because you know, like none of this is 100%, a hundred percent, but hopefully we can get. like 90 something percent. This is sort of the last dish.

0:35Unidentified speaker (UNKNOWN): Hey

0:36Unidentified speaker (speaker_1): everybody, it's the Quantum Bitcoin countdown today. Saving quantum vulnerable coins. There's over six million at risk. What can be done? Who will act or who should get involved in the clip? cleanup? Today's special guest, you know him as the co-inventor of Lightning Maelstrom Grantee. Talking about Mr. Taj Dryja. As always, I'm joined by my intrepid co-host Nathan Jaffe, head of media over at Starpware. Taj, you are opening up a can of worms here. Uh, you're out to take people's keys, redistribute them. What's going on? This video is made possible by Starkware. While others debate, Starkware is putting its top researchers on the quantum problem. You can learn more at Starkware.co.

1:24Unidentified speaker (speaker_0): Uh I tried to minimize that actually. But yes, I'm working on uh postponed recovery protocols for Bitcoin.

1:32Unidentified speaker (speaker_1): So where do we let's start with the size of the coin pool at play here? Are you looking for solutions across the stack? We mentioned, you know, there's differ various ages of keys, addresses, signature schemes. People have been knowing and following the podcast uh will know that signature schemes are a problem area that uh need to be addressed. So uh where does your work pick up and is there a particular area of focus when you think about all of the coins uh that may be of interest? to uh the quantum computer user.

2:04Unidentified speaker (speaker_0): Yeah. So I think a good way to look at what I'm working on is like hopefully it will ideally it will never be used. In that there's a bunch, this is sort of the last ditch in that there are a lot of layers before you need to use a recovery protocol that hopefully everyone will get to. So the first is hopefully a quantum computer never exists. Like maybe it just turns out to be a lot harder than we thought, and they never build one, and that'd be great. Then we don't have to do any of this. Um, but that's kind of risky, right? Because maybe they will come up with one. So like we should have a plan. And then the first plan is there will be a new signature scheme, a new um address type in Bitcoin. And it looks, it seems right now, at least all the people I've talked to, all the research is like shrinks. So it's like, okay, let's add. well, activate shrinks, which is a hash-based signature scheme, have a new address type, and try to get everyone to move over to that. Um and and then there's things like pay to taproot v2 where

3:01Unidentified speaker (UNKNOWN): it

3:01Unidentified speaker (speaker_0): shrinks, but you can still use elliptic curves until they get more iffy and stuff like that. Um so if everyone's moved, then you don't need a recovery scheme. The recovery scheme is only if

3:10Unidentified speaker (UNKNOWN): the

3:11Unidentified speaker (speaker_0): quantum computer shows up and people are like, oh shoot, there's a quantum computer. Um, and then there's lots of coins that have not migrated to a new address type that is quantum secure. And some I actually at the um meeting in Amsterdam recently, which wasn't supposed to be a quantum meeting, but discussion about. Um, there were some arguments that, like, well, maybe we shouldn't really have a recovery protocol, because if there is one, that will mean people can get completely complacent and they won't migrate to the new address scheme when that's available. Um and I think like, uh, okay, I get that to a certain extent, but really you should still have this. It's like, well, maybe we shouldn't have seat belts because people will be better drivers if they know that like instant death if you crash. It's like, yeah, okay, but you should still make seat belts and airbags. You know, like like I I get that there may be some incentives that move a different way, but um

4:03Unidentified speaker (UNKNOWN): I

4:03Unidentified speaker (speaker_0): think it's sort of inevitable that even if you have a new address scheme that's quantum secure, you have years to migrate, five, even ten years. And even if it's a majority of people that have moved to the new address type, you're still gonna have a lot of coins that haven't moved. Um Did just because if you look right now, there's still a lot of a lot of transactions that don't use SegWit, and SegWit is a decade old and saves fees. Um and where the quantum address types, quantum signatures are a bit larger and and less flexible. So I think there's gonna be people that don't migrate now. The idea of what I'm working on is like, okay, it's sort of the worst case scenario where quantum computer has showed up, it can now attack people. How can you save the coins that have not moved? Um and make sure that the only the rightful owners can move them after the quantum quantum computer is already here. With you know, and and so there's all these things like, well, what if you plan ahead and do this? It's like, no, no. The that would be great, and we hope that do that, but I'm only working on things where if

5:05Unidentified speaker (UNKNOWN): no one has

5:06Unidentified speaker (speaker_0): planned ahead, right? We're we're all I'm only focusing on people who have have not done anything. And I think that's still a

5:11Unidentified speaker (UNKNOWN): true.

5:12Unidentified speaker (speaker_0): One of the ideas with Bitcoin is you can put it on a razor, bury it in a you know, hole for ten years and come back and you still have your coins. And if during those ten years we did all this quantum stuff and it's like, no, you were supposed to move, and you're like, Well, I wasn't paying attention to Bitcoin for ten years. Like that I think that is a legitimate use case of I don't pay attention for ten years, I should be able to move it safely. Um

5:35Unidentified speaker (UNKNOWN): Yeah, much of the world that

5:37Unidentified speaker (speaker_3): much of the world that results Rizzo and I have been talking to uh is kind of you know, the world that's working on what I would call, you know, the contraception approach, you know, the world that's saying, let's get ready. And, you know, what we're talking about now is the emergency contracepts and the uh the morning after pill for the people who

5:55Unidentified speaker (UNKNOWN): had a

5:56Unidentified speaker (speaker_3): few many too drink, a few too many drinks or weren't paying attention. And you know, both are both are probably needed in the world. So yeah, this is all fascinating stuff. Now, I think Rizzo and I have had some really interesting conversations with Yehuda Lindel about this concept of, you know, he feels very strongly, and Rizzo can summarize who this is and the significance of the conversation, but he feels very strongly that people should be able to disappear, to lose interest. in Bitcoin, and that the community actually has an obligation to actually look after their assets or to ensure that their assets are not compromised. And I think for both Rizzo and myself, the conversation with Yehuda was kind of a a new perspective in terms of, you know, him saying actually not doing something is also a decision. When you don't do something, that's also taking a stance and maybe the community has some kind

6:56of obligation.

6:57Unidentified speaker (speaker_0): Yeah. I well, I don't want to Okay, I don't know about obligation because I'm like, well, hey, I'm I'm just working on this stuff, right? Like don't don't sue me. Um but at the same time, I I I guess the way I think about it is if you can, yeah, you should. Right. If there's if there are easy ways and I don't okay, not easy, but if there are pretty clear ways that don't have huge downsides, that it's like, look, we can do this. Here's the code, and this would save millions of bitcoins. It's like, oh yeah, we we should do that, right? Like it'd be kind of crazy to be like, no, no, we're not gonna do that. Like we have the code, it's not too hard. Yeah, maybe it takes an extra few megabytes here and there, but like we're just not going to screw you.

7:34Unidentified speaker (UNKNOWN): Well

7:35Unidentified speaker (speaker_1): well, let's let's start with like the worst case scenario version of this idea, because I think it's helpful to use analogies that are intuitive, right? I think like what people don't want is you know, okay, well, there's this key registry, let's just say it's owned and operated by a single operator. You know, you kind of go to them, you give them some version of uh, you know, key material that they can use to prove later, and then you know, there's like there's the quantum DMV, right? All of a sudden, you know, there's a flood that's gonna come to the town, and so you you'd go. go give your property to this, you know, uh guy who lives up on the hill and he's gonna make sure it's safe for you. And then, you know, there's this sort of elaborate, you know, process where you sort of have to appeal to authority uh to get it back. Let's just say it's the lost dog, you know, the guy's keeping trash of everybody's dogs after the flood, and it's just a cumbersome, it's bureaucratic, it's everything that we don't want in Bitcoin because in Bitcoin we want our users to be sovereign. we want them to have a direct relationship with the protocol. And I don't think we want an intermediary uh in the way uh of their ability to access uh that protocol. So for me, that's where it really breaks down, right? It's like, what is this intermediary? Is it a code? Is it a protocol? Is it an institution? Uh and so let maybe let's just start with this obviously bad idea. Uh, and and where are you taking it from that? And where's where's your good idea?

8:53Unidentified speaker (speaker_0): Sure. Uh so yes, I have heard things like that where it's like, oh, this company or whatever, they'll just figure out who has the coins. It's like, come on. I I think it's mostly people who are not super well versed in Bitcoin that would come up with these things. Because anyone working on Bitcoin for a long time is like, what? That's that's not how this works, right? There's there's no one in charge. There's no, you know, we can propose protocols and and software. but we're not gonna have a, you know, third party or some registry or something like that. Um and so, but fortunately, I guess what's what's interesting to me is working on this for the last, I don't know, year, there's a lot there's way better ways we can do it that are are quite minimal impact, that don't have any kind of registry, that don't have anything weird like that, where you just sort of because I guess the fundamental thing. is what would this registry even uh accept from a person? And whatever they're using to verify that someone owns a coin, can't that be just used by all the nodes in the network uh in a trustless, decentralized way? And it turns out that yeah, like a lot of the things that you would use to prove, well, I can't, I can't just sign in, anything, right? We're assuming that the elliptic curve signatures don't

10:03um we're assuming that the elliptic curve signatures don't work anymore because the quantum computer can find your, you know, get your public key, immediately find your private key and sign. So that doesn't work. But you still have a lot of uh I don't know what good word for them toe holes, hand holes, um secrets that the quantum computer can't find that you can use to show that you're there they're your coins. Um so for example, the the two biggest, and there's there's other little ones, but I think the two biggest to talk about, and they're interesting in different ways, are your public key. Um, so you have a in a lot of address types, you have a public key hash. And you haven't actually shown your public key. So your public key, it's weird because it's called public, but it's actually kind of hidden. And you can say, hey, I know my public key, and a quantum computer wouldn't, and no one else would, and I can spend based on that. The other option is some kind of seed phrase or deterministic key derivation path, where

11:05you can say, Well, maybe my public keys are known, my addresses are known, uh, the quantum computer can find my private keys, but they can't find my seed phrase. That's sort of up the stack that is not determined, you know, a quantum computer can't figure it out. So if I can use my seed phrase to show that look, these coins come from this seed phrase, and I have that seed phrase, and I can use the uh that seed phrase to then move the coins. So there's two, those are the two big categories, I guess. There's there's other little ones. Um, but those are the two big things that like you keep private, and because those are hash-based, right? A pub key hash or um BIP32 derivation involves hashes as well. The quantum computers can't reverse those hashes. Um and so there's still some sort of quantum secure way, quantum secure secrets that you still have. And you would need a new protocol to sort of prove that you have these quantum secure secrets and use those secrets to move those coins that are still there.

11:58Unidentified speaker (UNKNOWN): I

11:58Unidentified speaker (speaker_1): guess I'll say initially, I'm I'm a bit shocked to hear you say that the the the actual you know uh

12:03Unidentified speaker (UNKNOWN): you know

12:04Unidentified speaker (speaker_1): 24-word or whatever word private key is is a version of that because in a lot of uh maybe this is just mixing up with the post-cold card and the AI-related uh uh hacks. But um my understanding, or I think our understanding of the quantum computer, when you are breaking elliptic curve cryptography, what is exposed is the private key. Uh but you're saying that that was is somehow on unrelated to the 24-word seed phrase that there's that that maybe that is a misnomer that these are kind of both exposed in the quantum hack. I'm just curious if you could maybe elucidate. Yeah.

12:39Unidentified speaker (speaker_0): So the way the seed phrase works, and this is not super like this is a little bit hand-wavy, but basically correct. Um you take the hash of the seed phrase, right? So you have this seed phrase, these 24 words. you throw it into actually SHA 512, but same thing as SHA2P6. You throw it into a hash function, and then it spits out some random looking number. And then with BIP32 derivation, you actually do that a few times. So you know how there's like a derivation path where

13:03Unidentified speaker (UNKNOWN): it's a

13:04Unidentified speaker (speaker_0): servo 44 and so on. So each step you're doing another hash function. There's a couple other components involved in that you like take the hash and shop. chop it in half. But but for the purpose of what we're talking about, you're doing hashes. So it's you know five or six hash function outputs. And the quantum computer can't go back up. It can say, give it, you know, the quantum computer sees a public key and say, Ah, I can find the private key. And that's supposed to be one way, but the quantum computer breaks that. But what the quantum computer computer cannot do is it can't say, oh, here's the hash of something. I'm going to find what was the pre pre-image image or what was pre going into that hash function. Uh quantum computers can't do that. I mean, there is an ar like there is an algorithm that quantum computers can do called Grover's, which does make it faster to get a pre-image, but not fast enough. It's still very difficult. And there's actually been recent papers that show that Grover's is is kind of impractical practical and and not parallelizable. Um, so there's really not any worry. I mean maybe in theory in

14:04the thousand years or, you know, but it's but the actual thing we're worried about is um the public keys can be computed, you know, private keys can be computed by public keys. But pre-images from hashes is not at all the same thing. So because the seed phrase uses hashes, you can find the private key, but then from that private key you can't figure out the seed phrase. So the seed phrase still says secret. even when subject to a client.

14:29Unidentified speaker (speaker_1): And then on the public key side, I assume we're we're talking about this is a not exposed public key, is the other valuable path where I would have a fresh address that I used and maybe I funded, I would know the private key, but now this public key isn't exposed. Those But it sounds like I might have to give it to some, I might have to do something with it. somewhere. I won't say someone, but to something. Right.

14:54Unidentified speaker (UNKNOWN): So

14:55Unidentified speaker (speaker_0): with both of these. So I think what's useful to look is these are the two different sorts of I should come up with cool names for these, but these are the two like secrets that you can

15:03Unidentified speaker (UNKNOWN): use. Escape routes.

15:04Unidentified speaker (speaker_1): They're the two they're the two escape routes. Yeah,

15:06Unidentified speaker (speaker_0): the

15:07Unidentified speaker (UNKNOWN): two

15:07Unidentified speaker (speaker_0): escape routes and two angles. And the methods that that you then use to sort of prove this to the network um can be the same either way. right? So whether it's I have a seed phrase or I have a hidden public key, you can use you do need to prove them, but you can use the same methods to prove these things. And so there's different methods to do that, but there's like

15:28Unidentified speaker (UNKNOWN): I

15:28Unidentified speaker (speaker_0): guess the three, there's one called Dropkick that Conduition works on, and the one I call lifeboat, which is mine, these sort of commit reveal schemes where you commit to it in one transaction, then wait a while. and then reveal the secret. And that's a little um it feels weird because you actually end up revealing the secret at the end, revealing the pub key or revealing part of your something like your uh seed phrase. And that feels weird, but it's it's doable in a way that's secure enough. And then there's also like zero knowledge proofs where you say, oh, I'm gonna prove I know something about this seed phrase and that then it leads to this. um, or I prove I know my public key first. Um, and that doesn't actually reveal it. Those, so they're they're nicer in some ways, but they're quite um they're pretty big. And so that's sort of the trade-off there is that the zero knowledge proofs have the nice property of not revealing the actual secret, but are very heavy and very hard to verify and create. Whereas the other commit reveal schemes. are very small and fast, but they actually reveal the secret. I think it's I so I lean more towards the the commit reveal schemes because the assumption is it's okay revealing the secret because it's like, well, they were, you know, they were going to be lost anyway and I'm moving to something new. But there are downsides to that where like you kind of get one shot, where if you're like, okay, I have this hidden pub

16:48key and I need to use this to make this uh commit reveal to move my coins to a new place. You can't

16:54Unidentified speaker (UNKNOWN): do

16:54Unidentified speaker (speaker_0): it halfway. You can't say, Oh, actually, I want to change, oh, I want to you know, sign with a different fee or something. It's like a one-shot thing where once you start this process, you move to the new quantum secure thing, and that's it. So it's a little fragile compared to what people are used to, where you can, you know, today you can sign a thousand times with the same elliptic curve private key and you're all set. These are kind of one-off.

17:18Unidentified speaker (speaker_3): Yeah. It's interesting hearing you speaking about the ZK side of things because obviously working at Stock where I work with kind of, you know, very much the ZK, the ZK crowd, um, you know, and I hear, you know, talk of, you know, ZK proofs getting smaller, getting lighter, and also the, you know, kind of a pushback against what people sometimes cynically call, you know, disposable uh disposable cryptography, that kind of thing. Um yeah, how how do you view that kind of conversation?

17:49Unidentified speaker (speaker_0): I think so, in general, like yeah, these zero-knowledge proofs are really cool. Well, it's just they're if you assume that there's lots of people trying to do it and we're verifying the other it's larger and slower, fine. If there's a real benefit, then people would want to use this. If we can do something like that offline, I think you know we don't care about signing time or creation time. If it takes them an hour to sign, fine. It's that we worry about the verification time, where if it slows down the network work a lot. And I think the other thing people think about is to some extent it may be. new cryptographic assumptions. And it feels like when we're doing this quantum transition or whatever you want to call it, at least with the signature schemes and other things, it's very conservative because it's like, well, we don't know what's going to happen. We don't know if lattices are secure or isogenies or whatever. So we're just going to use hash-based signatures. And the commit reveal schemes are also very simple hash-based schemes. And so

18:51I think that's one of the reasons people like it. And myself as well. It's like, look, this is really simple. It's uh doesn't use anything that we're not already using in Bitcoin. And so I think that's what the interest is. But there's definitely also people working on zero-knowledge proof-based uh schemes

19:06Unidentified speaker (speaker_3): from

19:07Unidentified speaker (speaker_1): you. So look, we're going back to kind of the topology and the two escape routes. Uh, let's call them that. Uh, you know, uh, one thing you kept saying was that. the network, the network. So, okay, so is the idea here is that we're reusing parts of the cryptography that you as a Bitcoin hodler, you're already you already have the possession of these materials. But then the arbiter here is going to be the network itself. So we would have to we would have to teach the network something new, or what? So, yeah, what is this thing? Is this

19:37Unidentified speaker (UNKNOWN): going to

19:37Unidentified speaker (speaker_1): be a layer, it's layer two? or protocol. Like what is this thing?

19:40Unidentified speaker (speaker_0): No, I was I think I mean, I so there's different ways, and I've been working on it, but other people have different ideas. But I think a good way to do it is a soft fork, right? So you can do this as a soft fork, and maybe the soft fork is activated by people deciding. Maybe it's activated by a tripwire, I think is the name people use, where it's like you have a proof that the quantum computer has broken certain. things. Um so it could be, you know, it could be put into the code, sort of dormant, and say, well, when this happens, when you see a transaction with this kind of you know a signature from this key, you know the quantum computer has shown up because this key shouldn't be able to make the signature otherwise. Um and then it can activate. And then what it would activate is saying, uh you can't just sign a regular elliptic curve transaction. You can can't, like you. still need to, but you need additional data in order for this to be valid. Um, so that makes it a software because the the nodes, if you're running old software, won't know about this extra data. And they'll see it and they'll be like, I don't know, there's some op return in this transaction. Okay, I don't I don't care what that is. But the new nodes are gonna say, hey, when you're spending from an elliptic curve, you know, regul regular an old output that hasn't migrated yet. you need to also put an op return in your transaction. And that op return gives you me some proof about your seed phrase or your public key or something

20:58Unidentified speaker (UNKNOWN): like that.

20:59Unidentified speaker (speaker_0): Um, so it would be it would be doable as a soft fork where you know if if you have enough nodes that have upgraded, they will reject transactions and blocks that don't have this additional proof. Um, but what's nice is the old software can still see this, they still see the you know, if you if you aren't aware that the quantum computer has shown up, your software will still check the signatures as you will show. It is a little redundant. And some people have said, Chester, this is big enough that maybe it should be a hard fork. That you know you should you should just completely fork and have the different, and you don't even have the elliptic curve signature. And I think in this case, there's kind of an argument for that. Like it's such a big change that maybe you sort of want to force. people to upgrade their software. Um, I don't know. But but it is doable as a software. And that's sort of what we've always done in Bitcoin. So that's sort of the up the the mental like model I have for it. It's like, okay, it's soft work, you don't need everyone to upgrade, but um you uh you have new requirements for Or signing. And for the commit reveal protocols, you'd actually have two transactions. So you'd first have a

22:03commit transaction where you put some data in like an op return and it's like a hash of what you the transaction you want to make. You wait until that's confirmed. And then hours, days later, you reveal your actual transaction and it references that old op up return. Um the idea being that you don't want the attack because once you've made your actual transaction, sort of you have no more secret, right? So you can't just do it as a one-step process. You can't, well, it you could, in theory, with some zero-knowledge proof stuff, but with the um reveal protocols, if you reveal your pub key and that's all you had, then you don't have have you no more secret, right? So the idea is you use your secret first and then you reveal it later. Um, same with the seed phrase kind of stuff. You if if part of the process is revealing the secret, you want not to do it. And so you make some kind of con you make some kind of commitment first and reveal what it was in a remove your ones.

23:06Um, so that's that's the part that's a little weird um because it's sort of this two phase. You make like one transaction first and then a second later. Um, but it is secure. and it uses sort of the nice property of the blockchain where, like, okay, well, that's actually been confirmed for a day, so I can't get rid of that data.

23:22Unidentified speaker (speaker_1): And you're talking about these uh these conditions that you're imposing. I just want to clarify maybe for people who are following along, trying to help them kind of come up to understanding. It sounds like we're specifically talking here again about the coins that haven't migrated or that are in insecure cryptography today. uh that are secured with a you know uh lower quality address or a more at-risk address. And so it sounds like but but it sounds like there are options where we could impose some additional signing condition on on everyone in that pool who met that condition. And that sounds like something you're comfortable with that proposing that as an idea, even though essentially maybe someone could be. imposed on who didn't

24:05Unidentified speaker (UNKNOWN): opt in

24:05Unidentified speaker (speaker_1): to

24:06Unidentified speaker (speaker_0): right so the the question is and this is one of the big ones and this is why I think there's yeah what if you're imposing a new restriction on signing that someone cannot meet you're and then in that case you'd effectively be destroyed

24:20Unidentified speaker (UNKNOWN): and

24:20Unidentified speaker (speaker_0): some people say well can you make this opt-in and like I don't see any way you can make this opt-in because if the idea is the attacker you know, doesn't opt in to this really additional requirement. So and that's where I think the real big distinction between the pub key and the seed phrase. Pub key is you you know you cannot be destroying any coins by requiring some commit reveal for a pub

24:42Unidentified speaker (UNKNOWN): key.

24:42Unidentified speaker (speaker_0): Because in order to spend the coins, you must know the pub key. Because in order to spend the coins, you need to know your private key. And the private key, you can derive the pub key. You know, like every every normal transaction right now has a pub key in it. So you know saying, like, okay, I'm making this additional requirement that you have to first commit to your pub key and then reveal it. That's a hoop to jump through, and that's maybe annoying, and maybe people wouldn't like it, but fundamentally you can't argue that like I'm destroying anyone's coins, right? It's like, well, yeah, I'm I'm I'm putting a new you have I'm forcing people to maybe upgrade their wallet software, but they can do it. Like, there's no way someone can say, I can't do that. Even if you have a pre-signed transaction and you say, I don't actually know my private key. All I have is a pre-signed transaction someone gave me that is a valid transaction and haven't broadcast yet. You can still use a commit reveal protocol with that, because that pre-signed transaction. On the other hand, in some way, it's nicer because okay, so this is one of the one of the arguments of the pub key is look, we've been calling it pub keys for 20 years. They're public. And if now we're saying, oh, well, I hope you made it private because that's what now controls your coins, like that's really bad, right? Like I shared it, I told people the pub key, or my software did,

25:56or I didn't know I was supposed to keep my pub key secret. You never told me that. Totally valid. On the other hand, the seed phrase has doesn't have that problem. And if you know your seed phrase, yeah, you are the rightful owner of the Bitcoin, right? That's always been the case. You can sign for the coins, you control them completely, sure. The problem with seed phrase is we don't know which coins have a seed phrase, right? There's a bunch of so the seed phrase stuff was, or there's two things. There's BIP32 and

26:24Unidentified speaker (UNKNOWN): Yeah,

26:24Unidentified speaker (speaker_1): BIP32, yeah.

26:25Unidentified speaker (speaker_0): BIP39. So I I shouldn't I'm I'm simplifying a bit. What I'm actually talking about is BIP32 derivation, which is they're often used together, but they don't have to be. Um

26:37Unidentified speaker (UNKNOWN): so

26:38Unidentified speaker (speaker_0): Bib 32 was like 2012, 13?

26:41Unidentified speaker (UNKNOWN): 12,

26:42Unidentified speaker (speaker_1): yeah.

26:42Unidentified speaker (speaker_0): Something like that. And then BIP39, I feel, is a little like a year later or so. But anyway, most wallets now, and and you could use either, right? You don't if you have a VIP39 seed phrase and don't use VIP32, which is very huge, you'd still be able to use this. If you have VIP32 derivation but not a seed phrase, you could still use this. So generally they're used to better. Although like Bitcoin Core wallet doesn't support seed phrases but does support uh BIP32 bit 32 derivation. Anyway, either of these will will suffice for this kind of the problem is. there's some wallets that don't use either right and they're pretty rare at this point almost everything uses it but if you have some coins from 2013 14 like yeah bitcoin core didn't use this until around then so there's probably a bunch of coins that don't have a seed phrase or don't have anything like this um other things that would work is like brain wallet so like there's still sites like I mean I don't think it's a good idea. But if you're like, oh I want to make my private key just like hash

27:43of this poem I came up with, there are sites that will do that. You can go to like brainwallet.whatever and like it works. That would be sufficient for a commit reveal protocol because you know some secret about your private key. You know the pre-image of your private key and you can prove that. But there are undoubtedly a lot of coins that don't have any of this. right? That didn't use derivation paths. It was just a wallet that just came up with random numbers, saved that as the private key, and just saved that in the file. And you can't prove anything about that because they forgot

28:14Unidentified speaker (UNKNOWN): it.

28:14Unidentified speaker (speaker_0): Or even something where you did use a derivation path or a seed phrase, but you exported a private key and imported it into a wallet. So all your wallet knows is the private key. It doesn't know anything about how that private key was generated. And so you can't prove anything about it. So those those definitely exist. And that's why that one I think is they're they're both they're both problematic in different ways. Then that would be stolen if that person they shared it with has access to the quantum computer and can now use it to take the coins. And the the seed phrase one is like, well, someone didn't do the seed save phrase, right? Then they're stuck. and they can't move

28:52Unidentified speaker (UNKNOWN): it.

28:52Unidentified speaker (speaker_0): So

28:53Unidentified speaker (UNKNOWN): yeah,

28:54Unidentified speaker (speaker_0): they're both kind of messy. So they're the big problems, but no. Uh they're not they're not 100% safety things. Um, I yeah, but I feel like, hey, if you can save 90, 95%, like we should probably do, but but it it's a philosophical thing. It's like, well, wait, is it worth saving ninety, ninety? ninety-five percent of the coins if you know that you're gonna destroy the other five? Or should you just let them all be,

29:20Unidentified speaker (UNKNOWN): you

29:20Unidentified speaker (speaker_0): know, let let what happened happens. I think if if the numbers are pretty high, then you can be like, look, percent of coins here. And yeah, there's probably gonna be like one percent that end up being destroyed, but those probably were mostly abandoned anyway, and people probably lost the private keys for those anyway. So like, eh. it's probably worth it in the end. Um, but that's the big problem because these are not necessarily like technical arguments, they're kind of philosophical.

29:46Unidentified speaker (speaker_3): Yeah, if your family if your family fortune is within that 1%, and that is a life changing uh decision, you know. This was what was so fascinating about the conversation with you Yehuda Huda Lindell, and I'd urge listeners, viewers, if you've not watched that one, to really go and see it because it did become philosophical about, you know, where the responsibility lies. If there is responsibility, and I think Rizzo and I were talking afterwards that, like, you know, Bitcoin sees itself as very much an individualistic community, if that makes sense. Uh, and and this is kind of challenging some of our assumptions about that individualism and uh how we function. You know, this conversation is so fascinating, and I think we could go on forever. But what I do want to make sure that we get to is, you know, we wake up tomorrow morning um and find that there is a significant capable quantum computer. What would the scenario look like tomorrow morning? And what would you like the scenario to look like?

30:51if it happens in two or three years time and you have had time to do everything that you would like to do and have received support from those that you would like to cultivate as well?

31:05Unidentified speaker (speaker_0): Yeah. So if it happens soon, like Moonlight tomorrow, it's a lot worse. Um, but probably still worth doing. because we have the ideas that we don't have all the software and stuff, but we have enough software that we could probably pretty quickly throw stuff together. But oh, that's sort of the worst thing if you're like, oh, we're throwing stuff together real fast in Bitcoin because it's worth so much money and you don't want to screw it up. Um, but you could say, look, we're going to now enforce Force that we need these extra pieces of data. You need this commit reveal kind of thing. If you don't have a quantum secure signature scheme already or confirmed at the same time, it's m it's much worse. It doesn't work as well because the idea is like, okay, I need to make these commitments first. And then when I reveal it, I can move my coins. But if I don't have a way to make a commitment that's quantum secure, uh I'm sort of stuck. You can still do it. You can still use um like mining. So the only the only transactions you see today that don't use elliptic curve signatures are miners, you know, the Coinbase transactions, because there's no signature there. So you could say to a miner, hey, here are all these like pop returns I want to put in your Coinbase transaction, because those are like my commitments. Uh and that could work, but that's really ugly because now it's sort of this out of band: like, how do you pay

32:25the miners for this? Um so really you want a quantum secure signature scheme to already be active before you start using this. Or you activate them both at the same time, and then you can buy some quantum secure coins from someone and use that to move more coins. Um but yeah, it would be a huge mess if it happened tomorrow. Um so but but maybe still doable, do a but you know, it would be much higher losses, much more. theft, but hopefully somewhat salvageable. I think if you have a few years, um then it's still messy, but much better in that the first thing you do is you say, okay, here's the new signature scheme, and it's trained. And so we have a quantum secure hash-based signature scheme, and we also put in some kind of activation mechanism for this for the uh say, okay, if we have determined the quantum computer shows up, and that can be done programmatically by like if someone spends from one of these keys that we know were not made in a way that should be should have a private key. So for example, the X coordinate is the hash of something. Well, you shouldn't be able to have a private key for that. If someone can sign with that, then we know, okay,

33:35activate this commit reveal. And at that point, people, hopefully a lot of people would have moved to quantum safe signatures. If you haven't, you can hopefully buy some coins that have from someone else. And maybe there even would even be faucets. Maybe people would say, Look, I know people need a little bit of coins to m use these commit reveal protocols. So I'm and I know people are gonna try to rip me off, but I'm gonna make a faucet and it'll give you three 300 hundred sats or 500 sats and you do a captcha or whatever. And so try to, you know, just to be nice. And yeah, maybe you'll lose a few thousand dollars. But there's some nice people, and if it's a really big mess, I think I might run something like that, right? Or give coins to my friends, whatever. So you need to get it so that enough people have the quantum secure coins that they can make these the quantum

34:23Unidentified speaker (UNKNOWN): sequences.

34:24Unidentified speaker (speaker_0): When they make those transactions, they have like an op return and commit to the transaction they want to move their old coins with and reveal a few hours or days there. Um so I think it would be messy um but if you have a few years heads up then it's a lot cleaner and you you know need to update the software but you can update the software after the fact and I think we're already getting there in that, like, there's already a lot of work on the post quantum signature schemes, on the activation mechanisms. I'm working on the recovery mechanisms. So I think, yeah, if we have three years, it'll be still a mess, but uh a lot cleaner. And then it'll be like, here's what to do. Software's already there. Here now it's activated. Here you need to update your mask.

35:12Unidentified speaker (speaker_1): Yeah, what one are the of other undercurrents with that we've in had these? episodes is just

35:15Unidentified speaker (UNKNOWN): as a

35:16Unidentified speaker (speaker_1): hodler, you know, as a user of Bitcoin, it's hard to know what to do, right? Even as I've gotten more informed about quantum, um, it doesn't feel like I can actively do anything with this knowledge other than just be maybe a little bit more fearful about the potentialities. Uh and so, you know, there is this idea that

35:38Unidentified speaker (UNKNOWN): you know

35:38Unidentified speaker (speaker_1): we we've asked these users who own Bitcoin now,

35:41Unidentified speaker (UNKNOWN): they're

35:42Unidentified speaker (speaker_1): managing, you know, 24 word C phrases, they have them stamped in metal, they're using hardware wallets. Uh myself, I use a multi-signature hardware wallet setup. You know, it's uh there's all this investment in sort of like the actual user hodler behavior. Um and it does seem like what you're proposing, at least it seems like you're cognizant of that, that we don't we don't want to disrupt that. But there are scenarios, and it does seem like again, I don't think there's any scenario where some coins don't get left behind. The question though is in this new scenario, does hodling look the same? You know, are we gonna ever have the level of experience right now where uh you can just bury your 24 words and not think about it? Do you think the post-quantum reality do we get to some usability, or do you think the reality is just more complicated?

36:28Unidentified speaker (UNKNOWN): Like

36:28Unidentified speaker (speaker_1): own and hold Bitcoin going forward after a quantum computer?

36:32Unidentified speaker (speaker_0): So I think people like I'm not as worried about the hodler kind of, you

36:37Unidentified speaker (UNKNOWN): know,

36:37Unidentified speaker (speaker_0): because it's like I have a seed phrase buried in the backyard. Okay, cool. You have a seed phrase. Like that means you can recover with this mechanism. Um, or in a lot of people, like a lot of cases like that, you also have not reused. addresses. So, like, one one you can say is like don't reuse addresses, because that's good anyway, right? And like even if quantum computer never happens and all this is doesn't matter, great, you're helping privacy of the network and you're not reusing addresses. Cool. So that and if you have both, even better, then whichever system happens to be used, um, you'll be safe in two different ways. If you want to like proactively move, I bet in the next few years there will be a new address type softworked into Bitcoin and you can move to that. Um, but I think it's yeah, maybe you don't have to, right? If you've if you haven't raised addresses, your pub keys are hidden, and you have a seed phrase, I think you're kind of covered. And so like I think that it's pretty good. Like the only then question is well, does the network actually actually? activate these kind of commit reveal mechanisms? I hope they do. I think it will. I I think

37:45it's sort of weird to compare the two, where it's like, well, there's one where everything can just be stolen. And there's another set of software where no, there's like a borderly way for people to move their coins that ensures that they actually move them. Uh even if that other way has some sprinkling of theft and loss in it, it's like, well, 90 whatever percent here versus 0% here. It's like, yeah, I'm gonna stay on the that's the good chain, right? I don't want to go on the everything gets stolen by the quantum computer chain. Um and in some cases, the everything gets stolen chain like can't even press if the quantum computers are like everywhere and everyone can needs to just steal everything. Like, why would a miner even look at a memory? if they're like, I have like three million coins I can just take? And the miner doesn't care about the menu. They don't care about other miners. They're just trying to build the longest chain they can where they take those three million coins. And so it can be a huge mess there. Um, so I think it's it's likely we will, you know, have like it's likely that there will be software and people will want to run that software, and the miners will go onto on that to software because the alternative. is pretty bad. And so I think what the the disar you know the the argument then is there's these things where it's like, well, fundamentally, is it better for there to be theft or loss? Um, hopefully that

39:04can be minimized, right? We want

39:05Unidentified speaker (UNKNOWN): to

39:05Unidentified speaker (speaker_2): minimize.

39:06Unidentified speaker (speaker_0): But what's better? And like for me personally, I'm more okay with theft because. I'm not the one doing the theft. Right? I can say, look, I did everything I could, right? I've

39:20Unidentified speaker (UNKNOWN): no

39:20Unidentified speaker (speaker_2): one, no one's

39:21Unidentified speaker (speaker_0): you're like, if you're a hold lawyer, you're not are you paying me? Not really. I don't know. Do we have a contract? Do I have an obligation? Not really. I'm trying to help people out, but I did everything I could. I said, okay, I'm working on software. There's a new address type you can migrate to if you want. If you haven't, you can use your pub keys, you can use your seed phrase. But um I

39:41Unidentified speaker (UNKNOWN): did

39:41Unidentified speaker (speaker_0): I didn't steal like if someone if if I didn't protect it completely, I'm not the one who like got a clone computer and tried to steal your coins, right? But I may have been the one that wrote software that required something for you to spend your coins that you don't have. And you could say, Hey, you destroyed my coins because of the software that I wrote. And so that's kind of like, ah, I don't. want to do that because someone might get really mad. Um, I feel like, and they might get mad at me because I didn't prevent the theft, but it's like, well, look, is that really like I feel like someone can blame me more for destroying their coins if it's like they're stuck forever, versus someone stole them. Because, like, wouldn't you blame the person that stole them, not the person that didn't write the best software to prevent them? from getting stolen, I guess? So I lean more towards not destroying and allowing a little bit more theft, because

40:34Unidentified speaker (UNKNOWN): you

40:34Unidentified speaker (speaker_0): know, I'm not the one who would be doing the stealing, and so maybe they'll go after the person that actually stole them, hopefully. And I can see on chain. Um but yeah, but there there is sort of a trade-off Off there. And so one of the ideas that hasn't been discussed much at all, and then two people is you can maybe do phases, and you could you could have like this time component where you would say

40:59Unidentified speaker (UNKNOWN): the

40:59Unidentified speaker (speaker_0): first thing we do is quantum computer shows up. Okay, for the first let's say year, you need to use a seed phrase to move your coins. Um, and we know that some people don't have a seed phrase, and so their their coins are stuck for a year, and that's really bad. And that sucks. And we're sorry. And then after a year, you can use your public key. Right? You can reveal your public key in this commit reveal and move it that way. We know that that method is potentially susceptible to theft because you may have shared your public key with someone that we're not aware of. And that person may be able to steal your coins. But so that's why we have this one year, whatever period in the beginning where the the seed phrase is not. subject to theft because it's only the people that have the seed phrase that can move it. And you were pretty sure that you never shared it. If you did share it, well, you already shared your Bitcoin, right? Like that was part of the deal. And so we can have some time period where it's first seed phrase-based and then um pub key-based. And so it's and we need to tell everyone: like, hey, if you have a seed

41:59phrase and you shared your pub key with everyone. you really need to move it now. Cause if you wait a year or two, it might get stolen. So move it now. Like you, here's your window. So I think something like that might work, and that may cover be the best coverage you can get. Um, because sort of an either scenario, but you do have this time component. How long should the time component be? B Should it be a year? Should it be years? Um, and and it is still annoying because you're then protecting. potentially freezing coins for a year or two that otherwise could have moved, and people might not like that.

42:32Unidentified speaker (UNKNOWN): On

42:32Unidentified speaker (speaker_0): the other hand, it's like, well, you might have you hopefully had years and years where you could have migrated in it, and so it's well is an extra year gonna really be that bad. So there's there's ways like that. You but again, they're hard problems that are really nice mathematical engineering problems, right? They're kind of like well, if it's already been sitting there for six years, we figure an extra two years is okay. Like, well, where'd you get that number? Like, kind of make them up. Um so I think you can recover a block, and it's probably for most people, you know, it you're going to be safe, but there's gonna be and then the the I don't like calling it Satoshi coins because.

43:15Unidentified speaker (UNKNOWN): we

43:15Unidentified speaker (speaker_0): don't know if it's satoshi. I call it like two thousand nine era pay to pub keyputs. Those we're pretty sure there's no recovery mechanism you can do anything, right? 'Cause BIP thirty two, BIP thirty nine, none of this stuff existed. There's no hidden there's no pub key hashes. It's just like yeah, I don't know. Those I don't think uh and we're pretty sure those are abandoned, but

43:36Unidentified speaker (UNKNOWN): not

43:37Unidentified speaker (speaker_0): much.

43:37Unidentified speaker (UNKNOWN): But

43:37Unidentified speaker (speaker_1): you think those are there's you think that's impossible to create a recovery scheme for those early kind of first couple year coins.

43:44Unidentified speaker (speaker_0): Yeah, so there's one um uh Corey Fields came up with a fun idea. I don't know if it really is that usable, but um, I think you call it landmines, where you can say, well, what if someone with these 2009-year 2009 era coins made a commitment before the bonding quantum computer ship? showdown? And then they can reveal that commitment when you try to spend their coins. We have a new restriction where if you're trying to spend these old coins, you first have to put up a bounty, right? You put

44:13Unidentified speaker (UNKNOWN): up like we were

44:14Unidentified speaker (speaker_3): discussing this earlier today, Rizzo

44:16Unidentified speaker (UNKNOWN): and

44:16Unidentified speaker (speaker_3): I. It's a fascinating concept.

44:18Unidentified speaker (speaker_0): Yeah, and and it's like, yeah, maybe I don't. I mean, it it's an interesting idea. I think in practice, there's a lot of coins from 2009 that are abandoned because this Bitcoin was not worth any money. You could just download it, click mine, and then generates hundreds of coins, and then you just delete it. And yeah, a lot of that happened at the time. And so there's going there's a lot of these outputs that have never moved, right? They're just mining outputs um direct to PubKe hat to PubKe, PubKy, no hash. patch. And no one has this anymore. So there's no way to make commitments. There's no way to make commit reveal. You know, it's just they're abandoned. Um and so you so it's tricky because you could say, well, if they have a seed phrase, we'll let them spend it with a seed phrase. But we know that the seed phrases didn't exist, right? Like none of that technology existed then. So we're really pretty sure that there's nothing, right? And so then it's like, well, yeah, you say to destroy it, you say to have this landline thing. I don't know in practice if that would really do much, because people would probably just be able to steal it. Um, so those those are a big mess. But hopefully we can at least minimize it to just those coins, right? Where we're they either get stolen or destroyed. Um, and the vast majority of the coins that people still think of as their own coins

45:37and still have and hold. um can be safely moved. So I don't think it's uh, you know, none of this is a hundred percent, but hopefully we can get like 90 something percent.

45:47Unidentified speaker (UNKNOWN): I

45:48Unidentified speaker (speaker_1): guess we'll uh fascinating conversation, uh a lot to digest. Uh we'll we'll round it out. I gotta ask you about your quantum knowledge. You a couple times in this conversation, you know, we've chatted Added a little bit about quantum timelines and just the quantum computer. It's it's obviously a its own science. scientific discipline that's you know uh requires years of research. We've talked to people from that side of the aisle. It sounds like have you tried to kind of bring up your quantum expertise? Like, what has that process been like for you? It sounds like you're still skeptical that we might see a quantum computer.

46:17Unidentified speaker (speaker_0): Yeah, I so I've I've definitely read a lot of paper, you know, I've read papers from people with like, okay, here's Schorr's Schor's algorithm and quantum Fourier transforms and look. looked at and talked to people. I'm I am kind of skeptical because a lot of people, I don't know, they're like, Oh yeah, we're gonna have this in three years. We just need a bigger fridge. I'm like, come on. If it I yes, I'm not an expert, but if it was literally you just need a bigger fridge, really, like you've had you know, this this whole industry has had billions of dollars for decades. You just need to make a bigger fridge bridge, and then you're gonna have Shor Stor's source algorithm. I I I'm a bit skeptical of that. Um and I and there is a sort of like, you know, companies raise money by saying, Oh yeah, we're gonna have this soon, and they get a lot of research funding and stuff. And I think it's from what I've talked to people, it's very hard. I've talked to people at MIT who said, Oh yeah, we call our research quantum computing, but it's just physics research. And but you can you say it's quantum computing. And so we're like, this could be easy. used for quantum computing, probably. We're working and you know, researching how photons interact or something. We'll call it that and we get more funding. So I think there's definitely some of that at the same time. Like, so I'm not super worried that this is gonna happen really soon. At the same time, like, look, it might right, like, even if

47:31there's only a five percent 5% chance of this happening in the next 10 years, that's really scary, right? Like, well, if there was a five percent 5% chance that like, you know, an asteroid would crash in your house in the next few years, like, yeah, let's let's get some funding to like figure out how to prevent the asteroid. So I I still I think that like, you know, the fact that it's maybe not going to happen is good, right? Like, great. I buffalo this doesn't happen. But we should, but it's a credible enough threat, and enough people are worried about it, even that, yeah, we should have things to make. mitigate. Because a lot of it is even if I don't think it's a

48:05Unidentified speaker (UNKNOWN): you

48:06Unidentified speaker (speaker_0): know super likely to happen, other people do. And I know other people that are like, I'm not going to hold Bitcoin, I'm not going to use Bitcoin because of quantum computers. And it's like, oh, well, we should have stuff for those people, right? Even if I don't think it's gonna happen, I think it's pretty unlikely. We're, you know, preventing people from you. use It's weird to say preventing, but there are people who would want to use Bitcoin if it had a better, you know, quantum secure signatures and recovery and stuff and had a better you know way to do this. So like let's build that and then they can use it. And hopefully this never happens because there's nothing in quantum, there's nothing in this post-quantum, you know, signatures and stuff that's like better, right? Because if it was better, we would just adopt it today anyway. So like it's not good, right? Like it's like okay. we have SAS-based signatures.

48:52Unidentified speaker (UNKNOWN): All right,

48:52Unidentified speaker (speaker_0): they're bigger. You know, the main thing we're trying to do is mitigate and like minimize bad things today. And we've done a good job. I think there's looking at it today. This is like one or two years ago. Two years ago, I'd be

49:08Unidentified speaker (UNKNOWN): like,

49:08Unidentified speaker (speaker_0): oh, this is like and and talking to other um Bitcoin developers. If you talked about this two years ago, a lot of people were like dismissive and like didn't want to think about it because it was just like, well, if that happens, Bitcoin screwed. We're all dead anyway. You know, like, oh, it's like nuclear war, we're all dead anyway. And it's like, but if you talk to people now, they're like, oh yeah, no, it'd be a big mess. Right. It's sort of like we just even the signatures, it's like, oh yeah, we've got signatures that are like a few hundred bytes, right? And that are quantum secure, and you know, those are the stateful ones. But so there's there's a lot of progress just in the last two years. Um, and so it's like, yeah, it's it's worse if it happens, but it's definitely survivable. Um, and so I think it's worth working on because you can see that. And then the other, this is kind of a crazy No, no, thought. But maybe if you work on this a lot, you prevent the quantum computer from happening. In that I don't know that anyone's getting like research funding to steal bitcoins. That seems like it'd be hard, hard pitch for a like a VC, like, look, we're gonna build a quantum computer, we're gonna steal all the bitcoins. Um, but maybe, and if you say, look, we've got really good protocols, um, we've got this new signature scheme, we've got commit reveals protocols. protocol. So even if you have a quantum computer, you're not really gonna going to be able to steal the bitcoins. Um

50:28To some extent, that along with all this migration to post-quantum schemes for like the rest of the world, like the internet and TS TLS and stuff, you might have less interest in building a quantum computer, and there might be less funding and less research going into it. So I think I don't even so it's kind of a silly idea, but I honestly would not be surprised if 10, 15 years from now. 20 years from now, people we still don't have a quantum computer. And you talk to people and they're like, Yeah, well, we would have had it if you hadn't cut our funding. You know, if we had just had another few billion dollars to do research and make these crazy refrigerators and lasers and stuff, we would have had it. Because I feel like that's kind of what you see with like fusion research, where they're like, Yeah, we would have we, you know, we said we would have had it by 2010, and we would have if you hadn't cut our funding, but we didn't have enough money. And so it's like, well, that might happen with this as well. It's like a very difficult engineering problem, and it needs a lot of smart people to work on it. And maybe they'll say, like, we didn't have funding, and it is related to how well the world is prepared for my computer shows up. So that's that's sort of an argument for working on it, even if you don't think it's that likely to happen, or even if you think you're working. on it will make it less likely to happen. Still working. There

51:42Unidentified speaker (speaker_1): are some critics out there, folks, who say that the Bitcoin core developers aren't doing enough, aren't taking it seriously. It sounds like you disagree with that uh that statement.

51:52Unidentified speaker (speaker_0): There so it's it's not Bitcoin core developer. Like, there are people who don't want to work on this. Absolutely. There are people who work on Bitcoin, Bitcoin Core, where are who like this is stupid. it's not going to happen and dismissive. Sure. And there's also people who work who don't want to work on it because they don't even make a stance on whether it's going to happen or not. I know people who they're they work on Bitcoin because they really like elliptic curve cryptography. They like the elliptic curve math. And they're like, look, if a quantum computer happens, all this stuff I'm really interested in is gone. I don't want to work on hash based signatures. I don't want to work on. lattices. I like elliptic curves and this Schnorr signatures, all this stuff. And so if Bitcoin, if Bitcoin might continue in a post quantum world, but those people might not be interested. They'd go do some other thing in math that they're not interested in. So, like, okay, fine. Like, you know, you can work on different stuff. I would say there are a bunch of people working on this now. Um, so I don't know, I think it's like Chatham House rules. or whatever. I'm not gonna say people's names, but we were a bunch of people were at a conference in uh Amsterdam a few two weeks ago, and it wasn't supposed to be a post-quantum conference. It was just like Bitcoin research, it was called Bitcoin Research Week. It was just research on Bitcoin. It's been going for a few years, and people talk about you know research topics and math and

53:10whatever. It was not fully, and so there was a lot. of quant post quantum stuff at that conference two weeks ago, to the point where one person saying, okay, I'm gonna do a session on not post-quantum signatures. Like, you know, some people were sick of it and so they moved and sort of said it's too much post quantum stuff. So there's a lot of um, I don't know, half the discussion there was about post quantum and quantum transition. So there's a lot more activity on it now. Um certainly it's not everyone's everyone. Some people don't want to work on it and think it's not going to happen. But I I don't think you can still say, oh, no one's working on it or not enough people. There's there's a lot of activity. And I would not at all be surprised if the next software in Bitcoin was to activate a post quantum signature scheme. Because there's seem to be a lot of support for that. Because it's like, well, even if this never happens, you have a new signature scheme. We're pretty sure it's secure. Cool. People can use it.

54:04Unidentified speaker (speaker_1): Tadge, fascinating. As always, um, a lot to digest. We'll put everything of in the show notes. If you feel lost, uh, feel free to go back into the archive for the Quantum Bitcoin countdown. We've got interviews with some of the brightest minds out in the space. So Jonas, Nick, Jamison Lop, and others. Uh Taj, where can people find out more about your proposals?

54:25Unidentified speaker (speaker_0): Um I'm gonna I'm going to post on Delving Bitcoin in the next day or two. I have stuff on GitHub, just github.com slash eabat. And it's called lifeboat and life jacket and shrink ray. I didn't get into that, but there's the methods that you use, you know, there's the secrets, the methods you use to make these secrets let you move coins. Um so that's the main thing I'm I'm writing about. Um but I'll I'll post more on delving. And hopefully this will all be on like, I don't know, Twitter and stuff. People start ar it's weird. People aren't arguing about it yet. Um I I'm a little hesitant to like post it because I know people are gonna start arguing a lot and like, oh man. Uh but that's that's gonna be part of what happens.

55:03Unidentified speaker (speaker_3): Nothing wrong with starting an argument. It's it's fuels the discussion, gives visibility, and look, you know, month on month, you know, Rizzo and I is are seeing the interest in this stuff is growing and um, you know, this is a learning exercise that is important for all of us.

55:19Unidentified speaker (speaker_0): Yeah. So we're gonna start interviewing with people. Well, I'm gonna, you know, try to offer options and see what people what fun things people come up with. So

55:26Unidentified speaker (speaker_1): Well, we appreciate your work, sir. We hope you'll check out the cast and uh we'll have you along as uh things progress, uh and we'll see. Maybe a uh soft fork work

55:35Unidentified speaker (UNKNOWN): in

55:36Unidentified speaker (speaker_1): uh Bitcoin's future. Uh appreciate you.

55:38Unidentified speaker (UNKNOWN): See

55:38Unidentified speaker (speaker_1): you next

55:38Unidentified speaker (UNKNOWN): time.

55:39Unidentified speaker (speaker_0): Thanks. See ya.

This page was produced automatically: the audio was transcribed with nvidia/parakeet-tdt-0.6b-v3+sortformer-v2.1 and interpreted by Qwen3.6-35B-A3B-UD-IQ3_S.gguf. Transcription and interpretation can contain errors; always confirm against the original episode. Nothing here is investment advice.
All insights