Research
Timechain
Search Donate
Sponsors
Cake Wallet, private Bitcoin and Monero walletSideShift.ai, no-signup crypto exchangeOrangeRock, non-custodial bitcoin spot and perps tradingBraiins Hashpower, buy and sell Bitcoin hashrateLayer Two Labs, drivechains for Bitcoin
S16 · E61 · 2:41:12

S16 E61: Alan Szepieniec on Privacy, Scaling & Quantum Resistance

11 Dec 2025 · by Vlad Costea · 13 min read
Published at block 927,410
Cover art for S16 E61, Alan Szepieniec on Privacy, Scaling & Quantum Resistance
MP3 (221 MB)

The audio keeps playing while you browse the site.

Ask the Archive

Embed this episode

Paste this on your site or blog. It is a small player that streams from here: no trackers, no cookies, no ads.

Preview the player

This one ran an hour over schedule and the stream dropped three times, so a good part of the evening was me filling dead air with sponsor reads while Alan Szepieniec reconnected. When he was back on screen, he was the cryptographer and systems architect who co-founded Neptune Cash, a layer 1 built on ZK-STARKs and proof of work, and who earned a PhD in post-quantum cryptography and later did research at the Nervos Foundation. The first program he remembers writing, he told me, "was a toy cipher that would rotate bitwise all your characters by a secret amount."

I came into the conversation with a list of objections, mostly about Bitcoin, and Alan answered them from the angle of someone who bet on privacy early. He says he was "on record as early as 2011 stating that Bitcoin is not anonymous when people didn't understand the difference." At the same time, he admitted that in the beginning he looked at Bitcoin and thought "Bitcoin is not gold, therefore it will go nowhere." That is the sort of confession I enjoy from a guest, so we started there and kept going for two hours and forty minutes.

A traceable asset cannot settle debt

Alan's case for privacy is a legal one before it is a cypherpunk one. "privacy is a necessary feature of money because a traceable asset cannot settle debt," he said. His example: "if I pay you in Bitcoin, except those Bitcoins had been stolen two hops back, guess what? It's not my Bitcoins, legally speaking." He likes the car analogy for fungibility: "If I borrow your car and return a different car, you're going to be quite upset because it's not fungible. Money is supposed to be fungible."

I tried the gold comparison on him, since Bitcoiners reach for it constantly. He pointed out that "the stamp on gold is easy to remove," which is a trick you cannot pull on a coin whose history sits in a public ledger forever.

My own objection came from the miner side. Bitcoin, I argued, "will not be censorship resistant unless we're going to make it so that miners cannot distinguish between one transaction and the other." Alan took it further and said "Bitcoin was supposed to be independent of politics, but it really isn't independent of politics if the regulator can say you're not allowed to confirm these transactions." I added that European MiCA rules were already making Romanian shops drop Bitcoin payments, and I said so bluntly: "Well, we have the regulations and the adoption is going down."

His answer was to build around the problem instead of lobbying about it. The goal, he said, is to "make KYC obsolete by building in privacy on layer one," and he wants to "develop in complete abstraction of regulations and build something that works when it is regulated and works just as well when it's not regulated."

Rating Zcash, Monero and his own coin out of ten

Late in the show I asked him to score the privacy of the main contenders, and he answered with numbers. Zcash got "an 8 out of 10." Monero today is a 4, but with full-chain membership proofs he put it at "probably a 9 out of 10." Neptune Cash, his own project, he rated "6, I guess, or 7 out of 10."

His reasoning for the Monero score had to do with anonymity set size: "It doesn't matter that you are hiding perfectly behind all of history if you're only a small set of users." For Bitcoin privacy tools he made the same point, noting that with opt-in mixing "your anonymity set is limited to all the participants who choose to partake, which is by nature a lot smaller than all the participants of Bitcoin." Ideally, he said, "you wouldn't be able to opt out of privacy to begin with but unfortunately we cannot engineer that."

Neptune Cash goes the other direction: "all transactions are private by default." I asked about the trade-off in complexity, and Alan said plainly that "Neptune Cash would have been a lot easier to build if we hadn't built in privacy." He also mentioned that he is "more at home at MoneroKon than I do at your average Bitcoin conference," and that he gave Monero credit on supply too: "Monero's money supply, while not finite, is still predictable and that's a hell of a lot better than the fiat system that we have now."

I mentioned that I had interviewed Josh Swihart when Zcash was $30, long before it reached $700. Cypher Stack's people were also on for Bitcoin, Monero and Zcash privacy. Alan's ten-point scale was a more compact version of that whole debate.

What a STARK needs, and what a SNARK needs

Alan's technical case for ZK-STARKs rests on assumptions. "The only cryptographic tool that you need to build them is a secure hash function," he said, and "the T in STARK stands for transparent," meaning no trusted setup. SNARKs, as he describes them, lean on elliptic curve assumptions and often a ceremony. I mentioned that somebody "was able to eliminate the trusted setup from ZK-SNARKs," and Alan agreed there is now "a whole new family of SNARKs distinct from the ones with trusted setups." Then he drew the line again: with STARKs you can reach security that is provable rather than heuristic, whereas with the alternatives "you get heuristic security instead of provable security." He allowed that "there are trade-offs, but we're getting there, long story short."

That matters for his next argument, about scaling. "The point is that every block contains one proof and you don't care about any other witness data," he said. A new node can sync from the latest block and its proof instead of replaying history. "And I would argue we need to think of an architecture where history is irrelevant." If you push privacy off to a higher layer, his view is blunt: "If you don't have it on layer 1, then your layer 1 is going to explode."

I had Abdel on STARKs and Bitcoin maximalism earlier in the season. Alan was glad to see the technology maturing: after the trusted-setup debate, he said, "this sounds like a win-win scenario to me."

Drivechains, OP_CAT and an opcode called OP_STARK_VERIFY

Alan has a specific complaint about BIP-300. As written, it lets miners steal from a sidechain. His question was why Paul Sztorc had not adjusted the design to use newer proof technology: "And so I wonder why Paul didn't update his BIPs to make use of that technology." The idea is that a proof would make fraud impossible instead of merely unprofitable. Paul and I have gone through drivechains many times, including activating drivechains via CUSF, so I could report to Alan where that conversation stands.

On covenants, he does not want OP_CAT. "We don't want OP_CAT, we want something that's tailor-made to the task. We want OP_STARK_VERIFY instead." I pushed back from the other side: my instinct is "Maximizing what Bitcoin does, not trying to gatekeep what comes into Bitcoin." Alan's own estimate of the cost was matter-of-fact: "In principle, it's only one soft fork," he said, "but it will be a very, very, very, very big one," and then "a whole lot of engineering."

I also raised the other opinion in the room, which is that a Bitcoin full of proofs and sidechains would no longer be Bitcoin. Alan's thinking is that you can leave layer 1 alone and still reach the same place by opting into an architecture where verification is cheap.

Neptune Cash under the hood

Neptune Cash has a supply cap that Alan stated as "40 and a half million coins is the asymptotic limit on the coin supply." In the notes I took: blocks every 9.8 minutes, 8MB blocks, with 1MB of that for the proof. His elevator pitch for the project was "Private, scalable, post quantum secure."

The smart contract model is not Ethereum's. In his words, "Ethereum really is an operating system, whereas smart contracts on Neptune Cash are really just programs related to that particular UTXO." The language is "Turing complete. The only downside is that it's a new assembler language that is tailor-made to the STARK engine that we built." He added that the proving stack caused headaches of its own, since some of the tools "were not open source when we got started."

Mining is where Neptune Cash departs most from Bitcoin habits. Producing proofs is memory-hard and GPU-based, and Alan said it is "very difficult to do it with less than 40 gigabytes of memory." That also rules out hobby hardware: "because the Raspberry Pi doesn't have enough RAM to produce the proofs that you need to produce." I asked about ASICs, and he does not fear them. "In our eyes, ASICs are a good thing when the time is right to get there." The mining loop itself he summed up in three words, "Upgrading, composing, and guessing."

I got carried away comparing hardware. I compared a $13,000 AMD Threadripper to a $3,000 RTX 5090 and asked which one earns its keep on a job like this. Alan steered me back to the memory requirement, which is what decides who can participate.

The inflation bug and the genesis reset

Neptune Cash has had its bad moment. Over the summer the chain suffered "an undetectable inflation bug," and the fix was to reset from a new genesis block, with holders redeeming their old balances. I mentioned that I had heard a similar story on another show about a different coin, and Alan did not dispute the comparison.

On price, he was equally direct: "So the good news is that we are six times above the all-time low, the bad news is that we are still significantly below the latest fundraising price, which is life." He also said "the single most important factor in driving adoption is price," and followed with a line about team psychology: "the information that price drives adoption is just too depressing, and so we develop instead of looking at the price every day."

I asked about the early-access side of launches, and he said he does not want a special tier. "You don't want to taint the project by having this singular event where some people have special access to a non-market price." I teased him about the opposite failure, a coin so scrupulous that, as I put it, it is "so fair that nobody feels like they have any incentive to join it." Alan laughed it off and kept to his position.

His time horizon for his own role is short. "We do not plan to stick around forever," he said, and "Money is a thing that's too important to leave any one person or any group of people in control of."

Lightning, custodians and my 133 megabyte blunder

Alan surprised me on Lightning. "I think lightning is the most underrated technology since Bitcoin itself," he said, and "High-frequency trading is what the lightning network was built for." I came back with the whitepaper number: it "recommends 133 megabyte blocks just so that all the users of the Visa Network can open and close one channel per year." That is a figure I use whenever somebody tells me Lightning solves scaling.

The mistake in this segment was mine. I had gotten a technical detail wrong on air, he corrected me, and I said, "now I feel like an idiot." That is the part of live shows I like, because the audience watches a correction land in real time. I have had plenty of Lightning builders on, including Super Testnet on Monero versus Lightning privacy and Ben Arc on LNBits, and Alan's take sits comfortably alongside theirs.

Custodians were the other half of that segment. I do not like custodial wallets, since BlueWallet and others have shown how an app can disappear on its users, and I said I do not buy "some cypherpunk utopia where everyone is holding their own keys." Alan's answer was that custody does not have to mean trust. "I think the future is custodians deliver a service, but they have been made incapable of fraud through zero-knowledge proofs." He also added that incentives matter: "if 10% of the reserves are at stake, then they had better make sure that their auditing is up to scratch and that their proof servers don't go down."

I brought up the other risk I worry about on exchanges. Binance's market dominance is a single point of failure for the whole industry, and I think atomic swaps are the way out. Alan agreed that exchange concentration is a problem.

Satoshi's coins and the quantum question

The longest stretch was about quantum computers, and Alan is not worried about the usual story. "I don't see the national security applications of having a quantum computer precisely because everybody with half a brain will have migrated to post-quantum cryptography before that quantum computer can be built." For Bitcoin's old coins he sees a different problem. "I guess if quantum computers are built, then there is no way to distinguish Satoshi from a quantum computer. End of story." And so: "If he sells now, then people will think that quantum computers have arrived."

On freezing those coins he argued against it: "That sets a horrifying precedent that Bitcoin developers can freeze your funds." He described Neptune Cash as a fresh chain: "That's one of the reasons why we decided that this has to be a new layer one, so that we can actually have post-quantum security from the start." There is also Hunter Beast on BIP 360.

When I asked who Satoshi was, Alan said, "My money is on just an anonymous cypherpunk who probably passed away." We spent a while on culture, since I think "the development is downstream of the culture," and Alan more or less agreed: "The strength of a cryptocurrency stands or falls with the ideology of its adherents, of its users." He quoted the line about a republic, "A republic, ma'am, if you can keep it," and followed it with his own version: "If you do not exercise the right to free money, free not as in gratis, but as in free from oppression and free from restrictions, if you do not exercise that right, you will lose it."

His take on Bitcoin's economics was less gentle. "You need Bitcoin to not scale, because if Bitcoin scaled well, then what that means is it's easy to initiate transactions with low fees," he said, which means miners "have an incentive for their transactions to make it into the next block" only when blocks stay scarce. I mentioned ruefully that I once ran a wallet with a duress feature, as in Unstoppable Wallet: a decoy PIN opens a BIP-39 wallet holding $500 for anyone who is looking over your shoulder. Alan's reaction was that this was exactly the kind of practical privacy he wants coins to ship by default.

A bad trade and a bet I lost

The last part of the evening was lighter, and I asked Alan about leverage. Four years ago he took a leveraged Bitcoin trade and it went wrong, and he has stayed out of debt since. He told the story plainly, and I told mine the same way.

Then I asked the question I ask everyone near the end: would Neptune Cash be around by next year in the form he imagines? He answered a different question first, saying a flagship milestone is "probably not going to happen this year." I told him I had a note to myself to come back with "I told you so" if the timeline slipped, and I will, either way. We ended the call with the stream still holding, which after three dropped connections felt like a minor victory.

Listen to Alan Szepieniec on Spotify, Apple Podcasts, YouTube, and More

Time stamps:

  1. 00:00:00– Intro & Sponsors
  2. 00:02:30– Welcome Alan Szepieniec
  3. 00:03:02– Alan's Background in Cryptography & Libertarianism
  4. 00:05:15– Path to Bitcoin & Privacy Disillusionment
  5. 00:07:30– Privacy as Essential for Fungible Money
  6. 00:09:44– Network-Level Privacy & Censorship Resistance
  7. 00:12:15– Debate on Privacy Improvements in Bitcoin
  8. 00:14:01– Scalability Trade-offs with Recursive ZK Proofs
  9. 00:17:26– Lightning Network Privacy Limitations
  10. 00:22:39– Lightning's Potential for High-Frequency Trading
  11. 00:26:41– Custodial Risks & ZK-Proven Custodians
  12. 00:31:21– Sponsor: Unstoppable Wallet
  13. 00:33:17– Sponsor: Sideshift.ai
  14. 00:34:20– ZK-STARKs Explained & vs ZK-SNARKs
  15. 00:36:12– Quantum Resistance & Transparent Setup
  16. 00:41:56– Drivechains (BIP 300) & STARK Integration Potential
  17. 00:44:58– OP_CAT vs Custom Opcodes for STARK Verification
  18. 00:49:46– Why Coinjoins Aren't Sufficient for Privacy
  19. 00:56:27– Lightning's Unique Privacy Properties
  20. 00:56:50– Halo 2 & Removing Trusted Setup from SNARKs
  21. 00:58:44– Zcash L2 on Starknet
  22. 01:00:26– Why ZK-STARKs on Layer 1
  23. 01:03:20– Neptune Cash: Privacy, Scalability & Smart Contracts
  24. 01:07:33– Custom STARK Engine & Assembler Language
  25. 01:09:09– Proof-of-Work Design & GPU Mining
  26. 01:14:08– Inflation Bug & Network Relaunch
  27. 01:17:13– Decentralization Roadmap
  28. 01:19:00– Current Drawbacks: UX & Proof Generation
  29. 01:22:39– Node & Transaction Hardware Requirements
  30. 01:29:36– Quantum Computing Threats Discussion
  31. 01:33:01– Satoshi's Coins & Quantum Risks
  32. 01:36:38– Store of Value vs Medium of Exchange
  33. 01:43:13– Regulations Hurting Adoption
  34. 01:45:40– Monero Talk & Signatureless Transactions
  35. 01:47:58– Monero vs Bitcoin Communities
  36. 01:54:06– Adding Lightning Network to Neptune Cash
  37. 01:55:06– Lightning Network & Atomic Swaps Potential
  38. 01:56:48– Removing Centralized Exchanges via Bridges
  39. 01:57:17– Neptune Cash Lightning Network Plans
  40. 01:58:36– Prioritizing Faster Proving & Smart Contracts
  41. 02:00:36– Rapid Fire: Max Supply (40.5M after burn)
  42. 02:01:24– Block Time (9.8 minutes) & Difficulty Adjustment
  43. 02:01:56– Block Size (8 MB limit)
  44. 02:02:21– Elevator Pitch for Neptune Cash
  45. 02:03:17– Sidechain/Drivechain/Rollup on Bitcoin?
  46. 02:04:33– Privacy Comparison: Neptune vs Zcash vs Monero
  47. 02:05:38– Mutator Set vs Nullifier Set Explained
  48. 02:06:48– Privacy Ratings (Zcash 8/10, Monero 9/10 future, Neptune 6-7/10 + scalability bonus)
  49. 02:09:18– Scaling Ratings (Bitcoin 9/10, Neptune 8/10 future, Zcash/Monero 6/10)
  50. 02:12:24– Price Movement During Interview & Shitcoin Debate
  51. 02:14:54– Premine vs Fair Launch Discussion
  52. 02:17:35– Ethereum Premine Success & Grin vs Beam
  53. 02:19:10– Zcash Dev Fund Model
  54. 02:22:17– Building Circular Economies & Network Effects
  55. 02:27:04– Could Bitcoin Become Like Neptune Cash?
  56. 02:30:44– Connection Issues & Wrap-Up
  57. 02:33:05– Legacy UTXOs & Quantum Vulnerability
  58. 02:35:09– Where to Follow Neptune Cash & Alan
  59. 02:37:09– Is Satoshi an AI/Time Traveler?
  60. 02:38:46– Final Thanks & Goodbye

Verify this episode

This is the SHA-256 fingerprint of the MP3 as published here, timestamped in the Bitcoin blockchain with OpenTimestamps. If one byte of the file changes, the fingerprint changes.

SHA-256
d83eaff9bdb1f5f02ac9e32c3d124fb2f7d8bc9e871ccd51f7af551abd28ed45
Proof
2025-12-11_s16_e61_alan-szepieniec-bitcoin-takeover-podcast-neptune-cash-zk-stark-quantum-resistance-scaling-privacy-audio.mp3.ots · all episode hashes (proof)
Status
Anchored in Bitcoin block 974,312
Timestamped
After release. The episode came out on 11 Dec 2025, but this proof was made on 3 Nov 2026 and anchored in block 974,312, so it proves this exact file existed by 3 Nov 2026, not on the release date.
How to check it yourself
  1. Download the MP3 and the proof into the same folder.
  2. Run sha256sum 2025-12-11_s16_e61_alan-szepieniec-bitcoin-takeover-podcast-neptune-cash-zk-stark-quantum-resistance-scaling-privacy-audio.mp3 (on a Mac: shasum -a 256 2025-12-11_s16_e61_alan-szepieniec-bitcoin-takeover-podcast-neptune-cash-zk-stark-quantum-resistance-scaling-privacy-audio.mp3). The result must match the SHA-256 above.
  3. Install the OpenTimestamps client (pip install opentimestamps-client) and run ots verify 2025-12-11_s16_e61_alan-szepieniec-bitcoin-takeover-podcast-neptune-cash-zk-stark-quantum-resistance-scaling-privacy-audio.mp3.ots. It hashes the MP3 again and checks the proof against your own Bitcoin node, then prints the block and the date the file existed by. Without a node, ots info 2025-12-11_s16_e61_alan-szepieniec-bitcoin-takeover-podcast-neptune-cash-zk-stark-quantum-resistance-scaling-privacy-audio.mp3.ots shows the block height to look up in any block explorer.

Share: X · Email · Link

Related

Comments

Comments run on my own server. No trackers, no tracking cookies.

0:000:00