S16 E4: John Light on ZK Tech to Scale Bitcoin & Make It Private
The audio keeps playing while you browse the site.
Read the transcript13,473 words, every line links to the minute it was said
Ask this episode
The first time John Light was on my show, he did not say much. It was a four-way conversation with Super Testnet and Alexei Zamyatin that I published in January 2024, and he seemed rather shy. I knew there was more to him. I had seen him on stage at LABITCONF 2023, talking about zero-knowledge proofs on Bitcoin with the kind of depth that makes you want a longer conversation.
A year later I got it. John had gone full-time at Alpen Labs in April 2024 and was working on product for Strata, which he described as "a ZK optimistic roll-up on Bitcoin built using a cutting-edge technology based on BitVM2." A few days earlier I had asked his colleague Liam Eagen the same question about Alpen. Liam is the cryptographer. John is the product guy, so I wanted the version a user could follow.
We recorded live in mid-January 2025, the week of the Los Angeles fires and about 12 hours after Max Keiser, an adviser to El Salvador's president, announced a plan to get a Bitcoin node into every Salvadoran household. Both came up. So did Drivechains, the block size, Zcash versus Monero, Bitcoin maximalism and the first plug-and-play Bitcoin node he built a decade earlier.
BitVM2 and Permissionless Challenges
I started with the jargon. What is BitVM2, and how is it different from Robin Linus's original BitVM and from BitVMX, which Sergio Demian Lerner built for Rootstock?
John's answer was one phrase: "BitVM2 enables permissionless challenging." In BitVM1, participants run a pre-signing ceremony that defines every possible outcome of the protocol, then lock bitcoin into a Taproot address. If someone tries to withdraw in a way that breaks the rules, a challenge period starts, a bit like the justice transaction in Lightning. The catch is that only the people who were in the ceremony can challenge.
"In BitVM2, anybody can do that." Anyone watching Bitcoin with a copy of the rules can fund a challenge transaction and stop an invalid withdrawal. For a rollup bridge, that changes the security model. Operators front users their bitcoin on withdrawal and later claim reimbursement from the bridge address, and if their claim is invalid, anyone can block it.
He contrasted that with the federated bridges most people know. "They just use a multisig," he said, and if enough signers collude, "they can just do it. Nobody can stop them." BitVMX, BitVM1 and BitSNARK, in his description, all still have permissioned challenging. And none of this needs a soft fork. "BitVM1 and its spinoffs, as well as BitVM2, all work on Bitcoin today." Soft forks like OP_CAT or CTV would reduce interactivity and could allow permissionless withdrawals, which he called "pretty significant UX or even security-related improvements."
Citrea, Arbitrum and Who Was First
Citrea sponsored my show at the time, and they market themselves as the first ZK rollup on Bitcoin. John was polite but direct. "There are no BitVM-based roll-ups like live on Bitcoin right now. So, I think claiming to be the first is perhaps premature."
He described Strata and Citrea as very similar: both EVM-compatible rollups intending to use a BitVM2-based bridge. The difference, he suggested, might end up like Arbitrum versus Optimism, or zkSync Era versus Polygon zkEVM, where the gap only shows in implementation details. "I think it's too early to say right now like what exactly those differences are going to be once these systems go into production." If you want Citrea's own case, Orkun Kilic made it on the show in Season 15.
Sidechains, Rollups and Who Gets Paid
LayerTwo Labs also sponsored the show, and I told John how much I admire Paul Sztorc's push to make people think about what Bitcoin maximalism should mean. Then I asked him to explain rollups versus sidechains.
His definition rested on two things: consensus and data. A sidechain has its own consensus and handles its own data availability, so it can reorganize independently of Bitcoin. A rollup inherits consensus from Bitcoin and stores its data there, so "it's just as difficult to reorganize a Bitcoin rollup as it is to reorganize Bitcoin." Data availability, he stressed, is what lets users build the transaction that stops an invalid withdrawal. If a rollup state is committed but not published, users "are unable to produce a proof that they own coins," which could permanently freeze the bridge.
I asked whether Bitcoin miners earn anything from rollups. Yes, he said, from bridge transactions and from every batch posted to Bitcoin, and a rollup could even hand block production to miners. He advised against that last option for chains with lots of MEV, because the centralizing forces could leak into Bitcoin mining. Does a rollup need its own token? "It's optional. You can build a rollup and just use Bitcoin. And that's what we're doing with Strata."
I compared rollups to Counterparty and Omni. He agreed they share the habit of embedding data that other software interprets, but said rollups can use ZK to strip signatures and post only state differences. "It could be theoretically as small as 15 or 20 bytes per transaction," compared with about 200 bytes for a typical Bitcoin payment.
On Drivechains he was open-minded but elsewhere. "I think it's an experiment that's worth running," he said, mentioning BIP 300 and BIP 301 and Sergio Demian Lerner's own draft for hashrate escrows. His focus is native ZK verification, which would need a soft fork. "If the security of your coins is just as secure when you're opting into these new systems as it is like holding Bitcoin on the main chain today... that's as good as it gets. And that's what I want."
A Bigger Block, But Not a Plan
I asked about raising the block size so users can enter and exit Layer 2s on their own. John's answer was measured. A modest increase, "from here to let's say 4 million weight units to 5 million weight units or 6 million weight units," is probably fine after seven or eight years of better hardware and bandwidth. "But I don't think it's something that we can rely on perpetually to scale Bitcoin to the world."
He would rather exhaust the clever options first: CTV congestion trees for mass exits, and ZK techniques that let users migrate between Layer 2s without touching the base chain. "Dev harder, build solutions that don't require actually increasing the block size limit." Later he described leaning on block size increases alone as "a somewhat lazy way to deal with the problem," because bigger blocks make it harder to mine competitively and to run a full node.
A Bitcoin Pragmatist, Not a Maximalist
I asked what Bitcoin maximalism means to him. He uses Vitalik Buterin's 2014 definition from "On Bitcoin Maximalism," and noted that a definition some considered a straw man ended up being adopted by people who actually believed it, "almost like a self fulfilling prophecy."
He does not call himself a maximalist. "If there is a term that would apply to me, it's probably something like Bitcoin pragmatist, which is like, if Bitcoin is the best tool for the job, use Bitcoin. But if something else is the best tool for the job, use something else."
The altcoins he has actually used are Zcash for privacy and Ethereum for DeFi, including a loan through Maker and work on Aragon. When I asked why Zcash over Monero, he answered in three words: "It's more private." A shielded Zcash transaction hides you among everyone who has used the shielded pool, while Monero's ring gives you "15 other people maybe" per transaction. He admitted shielded transactions are still a couple of kilobytes. His bigger point was that altcoins are a temporary solution. "Eventually you're going to be able to do the same kind of stuff on Bitcoin. And at that point, what's the point of the altcoin?"
I went on a rant about Zcash being treated like Terra Luna or Bitconnect. John's reply was dry. "Zcash, you know, it does what it says on the tin. I mean, to be fair, Terra Luna did too." What he wants is to bring Zcash-level privacy to Bitcoin, and he pointed to Liam's Shielded CSV, which posts only a 64-byte commitment per transaction. Strata itself, he was clear, was fully public at the time: it runs Reth as an EVM rollup, though privacy protocols like Aztec or Railgun could be built on top later. (Our Zcash episode with Josh Swihart and Kris Nuttycombe is the place to go for the other side of that story.)
He was less interested in Litecoin's MWEB. "I don't think about it. No offense to that community." When I pushed, he questioned whether Mimblewimble implementations are really private, since watching the peer-to-peer network can reveal the transaction graph.
Why Everyone Builds for the EVM
I remembered the years when everyone wanted an Ethereum killer, so I asked how the EVM became the default. "I don't know if anyone would agree it's the best, but it's certainly the most popular." The reason is momentum: the first fully expressive smart contract VM, a huge developer community and contracts you can copy onto a new chain. "It's good enough."
On whether inscriptions inspired Bitcoin rollups, he said the ideas emerged around the same time. Once people realized a standard transaction could carry roughly a tenth of a block in a Taproot envelope, rollup builders started doing the math. As for timing, he expected trust-minimized EVM smart contracts on Bitcoin through BitVM2 "probably this year," and native ZK verification more like five to ten years away.
Fires, Multisig and Making Money Slow
I wrongly introduced him as someone who had worked on a hardware wallet. "I never worked on a hardware wallet project. I did work on a full node project." That project was BitSeed, which he described as "the first plug and play Bitcoin full node on the market." Jameson Lopp, he said, was one of their first customers. BitSeed started on a Raspberry Pi-style board and moved to an Intel mini PC, and John said he still runs a hand-built node today. He was happy to see Umbrel and others carry the idea forward.
With the Los Angeles fires fresh, I asked how far he would go to secure coins. He refused to give one-size-fits-all advice, then offered a rule: "multisig with geographically distributed backups is the gold standard." He also said something I keep thinking about. "I would certainly err on the side of making it difficult to access your coins." Time delays and friction are features the legacy system learned the hard way. "We should make money slow. That's good for security."
On the El Salvador node plan and government Bitcoin reserves, he did not hedge. "I don't think it's right for governments to buy Bitcoin or buy Bitcoin nodes and put them into people's houses." He granted that a reserve might be bullish for price but said governments should "give that money back to the people they took it from, the taxpayers."
If Bitcoin Ossifies
My last real question was a pessimistic one. What if Bitcoin ossifies and we never get the soft forks? John did not think it would happen, because expanding Bitcoin's capabilities makes it more valuable, and "things that are good for Bitcoin make Bitcoin more valuable." Changes will come "on a Bitcoin time scale, but they will happen." And if not? "I guess we have to live with BitVM2 and altcoins."
I complained that Lightning had disappointed and that new Layer 2s have cheerleaders but no organic demand. He defended Lightning firmly. "I think Lightning is an incredible achievement. It is the only trust minimized L2 that operates at scale right now." He uses it regularly and said he had not had a failed payment in a long time. He also sketched a future where a developer builds something on Bitcoin with ZK, solves a real problem and onboards people who "maybe don't even realize that they're Bitcoiners."
I closed by complaining that OP_CAT debates had become tribal, more about the Taproot Wizards than about concatenating two pieces of data. John was more relaxed about it. Some people want the technical weeds and most just want working products. "I wish people would like more people would get involved in the technical discussions... But, you know, it's not for everybody."
For a guy who barely spoke in his first appearance, he turned out to be one of the clearest explainers of BitVM2 I have had on the show. Alpen's own David Seroy came on later in Season 16, and the full hand-corrected transcript of this episode is on the site.
Listen to John Light on Spotify, Apple Podcasts, YouTube, and More
Time stamps:
- 00:00:50– Introducing John Light
- 00:01:35– John's Work at Alpen Labs
- 00:02:35– Strata: A New zk Optimistic Rollup
- 00:02:52– Understanding BitVM
- 00:03:14– BitVM 1 vs BitVM 2
- 00:05:03– Challenge and Response Mechanism to Prevent Cheating
- 00:06:19– BitVM vs Lightning Network
- 00:07:26– Security of BitVM 2
- 00:13:01– Soft Forks and BitVM
- 00:14:06– Potential Improvements from Soft Forks
- 00:15:58– Citrea vs. Strata ZK Rollup Tech
- 00:17:46– Side Chains vs. Rollups
- 00:20:44– Data Availability in Rollups
- 00:23:19– Transaction Fees for Bitcoin Miners
- 00:26:25– Token Requirement for Rollups
- 00:27:06– Comparing Rollups to Counterparty
- 00:29:35– Drivechains and Their Potential
- 00:30:49– Focus on Rollups
- 00:31:42– Data Availability Challenges
- 00:32:52– Increasing Bitcoin Block Size
- 00:33:19– Importance of Data Availability
- 00:36:02– Block Size Limit Increase
- 00:37:46– Layer Two Solutions
- 00:40:19– Bitcoin Maximalism Defined
- 00:41:33– Vitalik's Definition of Maximalism
- 00:46:25– Personal Views on Altcoins
- 00:47:51– Zcash vs. Monero
- 00:50:36– Zcash's Privacy Advantages
- 00:51:07– Temporary Nature of Altcoins
- 00:52:32– Zcash as a Scientific Experiment
- 00:56:43– Consolidation of Currencies
- 00:57:35– Zcash's ZK Technology in Bitcoin
- 00:58:15– Zcash vs. Shielded CSV
- 00:59:44– Privacy in Rollups
- 01:00:50– EVM’s Popularity
- 01:04:00– Inspiration from Bitcoin Inscriptions
- 01:17:37– Security of Bitcoin Holdings
- 01:22:01– Full Node Projects and User Adoption
- 01:23:35– BitSeed's Hardware Evolution
- 01:26:31– Bitcoin Node Legislation
- 01:27:08– Government Involvement in Bitcoin
- 01:27:49– Concerns about Strategic Bitcoin Reserves
- 01:29:31– Natural Evolution of Government Bitcoin Holdings
- 01:30:28– Future of Bitcoin Development
- 01:31:01– Incentives for Bitcoin Expansion
- 01:32:36– Challenges of Bitcoin Solutions
- 01:34:02– Product Management in Bitcoin Development
- 01:36:16– User Enthusiasm for Layer Twos
- 01:37:25– Potential for New Bitcoin Solutions
- 01:40:37– Cultural Shift in Bitcoin Discussions
- 01:42:10– Engagement in Technical Discussions
- 01:42:53– Following John Light's Work
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
7929880db5cee36800334972dc2e1f24b47d830ee3e56839c73d4b84b88650cd- Proof
- 2025-01-16_s16_e4_john_light_on_zk_rollups_bitvm_alpen_labs_strata_-_16.01.2025_18.07.mp3.ots · all episode hashes (proof)
- Status
- Anchored in Bitcoin block 974,312
- Timestamped
- After release. The episode came out on 16 Jan 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
- Download the MP3 and the proof into the same folder.
- Run
sha256sum 2025-01-16_s16_e4_john_light_on_zk_rollups_bitvm_alpen_labs_strata_-_16.01.2025_18.07.mp3(on a Mac:shasum -a 256 2025-01-16_s16_e4_john_light_on_zk_rollups_bitvm_alpen_labs_strata_-_16.01.2025_18.07.mp3). The result must match the SHA-256 above. - Install the OpenTimestamps client (
pip install opentimestamps-client) and runots verify 2025-01-16_s16_e4_john_light_on_zk_rollups_bitvm_alpen_labs_strata_-_16.01.2025_18.07.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 2025-01-16_s16_e4_john_light_on_zk_rollups_bitvm_alpen_labs_strata_-_16.01.2025_18.07.mp3.otsshows the block height to look up in any block explorer.
Related
S15 E6: John Light, Super Testnet & Alexei Zamyatin on Improving Bitcoin
Bitcoin is still a young project that’s bound for disruption. While the money layer is supposed to never change, the advancements…
Preview: Bitcoin Takeover Podcast Season 15
Season 15 of the Bitcoin Takeover podcast marks two significant milestones: the 5th anniversary of the show, as well as the 200th…
S16 E27: David Seroy on Alpen Labs & ZK Rollups on Bitcoin
David Seroy of Alpen Labs on ZK rollups without federated custody, BitVM bridges, a Bitcoin devnet with mainnet aimed at Q4, and…
Comments
Comments run on my own server. No trackers, no tracking cookies.