2021-06-16 · timestamps by Lex Fridman
AMA 06/16/2021
- make the case for Cardano's peer reviewed, Haskell based approach to consensus (Ouroboros) and smart contracts (Plutus) as safer than move fast and break things engineering
- frame three generations of cryptocurrency (Bitcoin, Ethereum, Cardano) and argue a third generation chain needs built in governance and self evolution to survive at billions of users scale
- champion Cardano's Voltaire governance rollout, Catalyst, treasury, and constitution as decentralizing decision making away from IOG itself
- argue crypto's biggest real world impact will be in Africa and other developing markets, leapfrogging broken legacy financial and identity systems, citing the Ethiopia digital identity deal
- reflect on power, fame, and the burden of leadership, defend Satoshi Nakamoto's anonymity, and push back on price obsessed number go up culture
- share personal life in depth: daily routines, health experiments like extended fasting, video game projects, and philosophical musings on meaning, mortality, and simulation theory
285 entries
Gets introduced as a mathematician, Cardano founder, and Ethereum co-founder who also farms bison and mushrooms in Colorado and once posed with a hawk in Mongolia meeting the country's president.
Wonders whether the universe is a self-learning computational system, citing MSR's Autodidactic Universe paper and comparing humanity's position to being trapped inside a Minecraft simulation that cannot see outside itself.
Finds Stephen Wolfram's approach of letting simple rules evolve into complexity compelling, comparing it to how the Santa Fe Institute models economics and how AI systems learn from simple starting conditions.
Announces a partnership with Stephen Wolfram to auction his computational universes as NFTs on Cardano.
Reflects that simple deterministic rules can be beautifully complex yet still impossible to perfectly predict, unlike the fantasy of Laplace's demon.
Names Bertrand Russell and Saul Kripke as favorite philosophers for their work driving precision in formal language and the nature of truth.
Traces Bertrand Russell's career from Principia Mathematica through Godel and Turing's incompleteness results, which forced him to reinvent himself as an ethicist and pacifist writer.
Saul Kripke as a teenage logic prodigy who later resolved the truth paradoxes left unsettled by Tarski, calling him the Galois of philosophy.
Hopes new tools can bridge the gap between the formal world computers live in and the natural language world humans live in.
Digresses about Lojban, an unambiguous constructed spoken language equivalent to first-order predicate calculus, joking about tweeting and writing poetry in it.
Considers whether love, art, and creativity can ever be formalized, framing human connection and impressing others as an optimization problem for maximizing subjectively rich experience.
Jokes that fake it till you make it is his view on love when asked about the hologram relationship between Joi and K in Blade Runner 2049.
Distinguishes software engineering, which is obsessed with user requirements, from theoretical computer science, which gives guard rails like termination or no-overflow guarantees rather than perfect programs.
Cryptocurrency systems must decide what absolutely cannot fail, like funds and privacy, versus what can fail, like temporary latency, and apply the heaviest formal tools only where it matters.
Cryptocurrency protocols run in a far more hostile environment than a pristine data center, on ordinary phones with unreliable connectivity worldwide, demanding new classes of protocols and testing.
Criticizes perverse incentives in crypto where builders get paid up front and rush to market, leaving users to suffer when bugs surface months later.
Argues distributed systems are game theoretic at their core and that the assumption of a rational actor rarely holds since real people behave irrationally.
Cites the Cuban missile crisis and the 1983 Soviet false alarm as evidence that humanity is bad at building rational-actor systems, arguing designs should instead aim to be self-correcting like an immune system or a healing rainforest.
Reflects on COVID as an unplanned real-world experiment in monetary policy, where developed nations printed money while over 100 million people in developing nations faced acute starvation.
Compares the crypto market to a Darwinian experiment with over 8,000 competing monetary policies, where Dogecoin thrives against expectations while a strong team like Algorand's underperforms on market adoption.
Argues that individual project outcomes matter less than whether the overall crypto system converges over time toward something genuinely useful to society.
Recommends Python or JavaScript for beginners learning to program, noting even MIT switched its introductory course from Lisp to Python.
Cardano's choice of functional programming to minimize the semantic gap between rigorous academic papers and running code, at the cost of weaker Windows support and raw speed.
Haskell was chosen as Cardano's version one language because researchers who write papers like Ouroboros can also translate them into a verifiable formal specification and running code.
Cites the Mantis Ethereum client built in Scala at roughly 15,000 lines versus Bitcoin's 120,000 to 150,000 lines of C++ as proof that functional code stays far more concise and maintainable.
Frames the language choice as a Goldilocks problem between overly academic proof languages like Idris or Isabelle and pragmatic hybrids like Scala or Clojure, with Haskell landing in the middle.
Recommends Scala 3 as the language to build a career on today, praising Martin Odersky's work and Netflix's use of Scala for its microservice architecture.
Declares that Rust has essentially replaced the need for C and C++, alongside Go, which Google built specifically to get away from C.
Answers a Reddit question on why developers should build on Cardano by pointing to PlutusFest, an annual developer conference planned around the University of Wyoming hackathon and the Goguen summit.
Plutus Core is a compilation target, so languages like Scala or TypeScript could in principle compile to it, reflecting his belief in separating execution environments across chains.
Work with Grigore Rosu's Runtime Verification team at the University of Illinois to bring LLVM-based execution onto Cardano, using an island, ocean, and pond analogy for restrictive versus expressive chains.
Says that if the Ethereum Virtual Machine becomes the industry standard, Cardano will support it as an interoperable execution environment while keeping its own main chain conservative.
Defines Plutus as a domain-specific language built on Haskell and argues a smart contract is usually a small sub-module of an application rather than the whole app, using an in-game currency example.
Plutus aims for deterministic, locally testable resource costs for small to mid-size programs, contrasting it with Ethereum's global mutable account state.
Recalls Bitcoin's 2010 value overflow incident, which created billions of unintended Bitcoin, as a cautionary tale about scripting languages blowing up when developers try to extend them.
Promises a CryptoKitties-style demo by August showing the full round trip between an off-chain game interface and low-fee, low-latency on-chain Plutus contracts.
Defines a blockchain as simply a timestamped, immutable, auditable ledger, with cryptocurrency as just one application that runs on top of it.
Uses Syria's collapse and ISIS rewriting land ownership records as a cautionary example of why traditional government-controlled ledgers are brittle to political upheaval.
Imagines a tax system where billionaire tax records are transparent by default like the ProPublica leak, and describes cross-border medical record access as another blockchain use case.
Cryptocurrencies can carry any monetary policy, inflationary, deflationary, or demurrage, and can sit anywhere on a transparency spectrum from Bitcoin's openness to Zcash's privacy.
Breaks all consensus algorithms, from proof of work and proof of stake to Paxos and Raft, into the same three steps: deciding who is in charge, making the block, and getting the network to accept it.
Distinguishes proof of stake, where the resource pool deciding who is in charge is a token making it a plutocratic system, from proof of work where the resource is raw computation.
Walks through the three-step consensus model: pick a leader to order transactions, make the block, then have the network validate and accept it.
Compares proof-of-work's cult-like ideology to proof-of-stake's plutocratic weakness, where more ownership means more control over the system.
Multi-resource consensus systems that blend several selection resources, citing Peercoin's 2011 hybrid proof-of-work and proof-of-stake design.
Summarizes the ledger's core job of ordering transactions immutably, pointing to Lamport's foundational 1970s distributed systems paper.
Breaks down proof of work as a lottery-ticket system for computing devices, covering ASIC-resistant versus specialized hardware and sharded protocols like Prism.
Proof of stake as a synthetic resource that still requires making and validating blocks, unlike proof of work's physical resource.
Proof of work's huge energy cost drives centralization into a handful of mining operations, comparing its energy use to the nation of Switzerland.
Argues proof of stake gets more decentralized over time as token distribution widens, comparing it to Bill Gates's declining Microsoft stake, and describe Cardano's k parameter for tuning stake pool saturation.
Favor virtual resources over physical ones for security, since stake can redeploy instantly across jurisdictions, unlike miners fleeing China's mining crackdown.
Recalls that early proof-of-stake coins like Peercoin and NXT suffered grinding attacks and nothing-at-stake problems before the GKL paper formalized blockchain security properties.
Credits proof of work's genesis bootstrap property, letting new nodes trust the heaviest chain, until Ouroboros Genesis solved the same problem for proof of stake in 2018.
Flags proof of work's winner-take-all flaw, where a miner with 51 percent can profit by destroying and shorting one chain while mining a competing one, a Goldfinger attack.
Advocate incentivizing multiple resources, including paying for network relay bandwidth, so the system doesn't centralize around its weakest link.
Contrasts Cardano's approach with Ethereum's on the philosophical split with Vitalik: his team tries to solve everything at once while IOG's academics decompose the problem into a provable chain of steps.
Traces the Ouroboros protocol lineage from the GKL model through Ouroboros Classic in 2017 to Praos, each peer-reviewed at top venues like Eurocrypt, Crypto, and CCS.
The shift from proving theoretical feasibility to practical engineering questions like clock assumptions and random number generation.
Solving each harder consensus problem required different kinds of cryptographers, moving from pure mathematicians to applied engineers as the protocol matured.
Layer an economic delegation model onto Ouroboros Praos, comparing the rise of specialized stake pools to biological cell specialization.
Walks through how stake pool registration, on-chain certificates, and wallet delegation work, with pools charging an operating fee on rewards.
Details the k parameter's diminishing-returns curve that caps pool saturation and pushes operators to split into multiple pools.
Credits hiring Oxford algorithmic game theorist Elias Koutsoupias to model stake pool incentive curves, though he says the game theory of stake pools is still an ongoing, iterative problem, not fully solved.
IOG's expanding academic research labs in Edinburgh, Tokyo Tech, Wyoming, and Athens tackling interdisciplinary problems across crypto, systems, and economics.
Cites nice-to-have refinements like the Wormhole Chronos internal clock protocol and University of Illinois work moving leader election from epoch-based to block-by-block.
Value rigorous published research for creating a shared academic language that invites outside researchers to attack or build on Cardano's protocols.
Defines Cardano as a financial operating system meant to give everyone in the world equal access, not just people in rich countries.
Admits Cardano isn't the final solution and needs decades more evolution, built with a self-funding treasury and on-chain voting from the start.
Calls for a Wi-Fi or Bluetooth moment for blockchain interoperability, since chains are currently blind, deaf, and dumb to each other, and stresses the system must scale to billions.
Frames Cardano as a first approximation of a self-evolving system that, done right, doesn't need founders, citing the lesson of Satoshi's disappearance.
Points to Cardano's decentralized academic base of over 30 contributors across 105 papers, expected to grow to hundreds of scientists worldwide.
Entering Ethiopia to bring digital identity to 5 million people as the foundation for economic agency and future on-chain applications.
The cascading disruption strategy of being the first domino, letting adoption spread through self-interest rather than charity.
Contrasts Cardano's bottom-up, self-sovereign identity approach with China's top-down digital yuan and social credit system.
Unpack why the financial part of financial operating system is essential, since managing and trading resources intelligently is fundamental to how people survive and interact.
Frames the industry's purpose as killing the middleman, turning any invited third party into a value-add rather than a necessity.
Highlights the financial stem cell idea, where the same token architecture can represent a national CBDC or a CryptoKitty.
Recaps the three-generations framework, with Bitcoin as the first generation solving trustless value transfer as a single-purpose system.
Cast Ethereum as the second generation, bolting programmability onto a blockchain the way JavaScript transformed static web pages.
Programmability brought ICOs, DeFi, and NFTs but also congestion, high fees, and systems too big to coordinate informally.
Uses Vint Cerf and the early internet's simplicity to illustrate how billion-person systems lose the ability to coordinate change informally.
Defines third-generation cryptocurrency as needing self-governance at billions-of-users scale, alongside being better, faster, and cheaper.
Survey third-generation competitors like Tezos, Algorand, ICP, and Polkadot, each optimizing a different blend of scalability, interoperability, and sustainability.
Rejects a winner-take-all future for blockchains, comparing coexisting protocols to Google, Apple, Microsoft, and Facebook, or to how TCP/IP simply became shared infrastructure.
Compares early blockchain demos to Xerox PARC's GUI and networked computing, proofs of concept that took decades to become real products.
Both Bitcoin and Ethereum are self-evolving systems that could in theory grow into third-generation systems, with Ethereum 2 as Vitalik's own upgrade path.
Criticizes Bitcoin as slow, unprogrammable mainframe-era technology kept alive mainly by sunk investment, like COBOL.
Lists concrete upgrades Bitcoin could make without compromising security, like faster consensus tricks from Prism, Blockstream's Simplicity language, or Cardano's extended UTXO model.
Traces Vitalik's path from a frustrated Bitcoin color-coins and Mastercoin developer to founding Ethereum after constant pushback from Bitcoin's core developers.
Predicts Ethereum would beat Bitcoin nine times out of ten in a head-to-head fight, thanks to its culture of embracing upgrades.
Argues Bitcoin has no built-in incentive to evolve, unlike user-driven platforms like Android and the App Store.
Recounts getting a toilet-paper roll marked shitcoin thrown at him at a Miami Bitcoin conference, questioning Bitcoin maximalist culture.
Defines Cardano's success as the poorest person in the world getting the same fair access as the richest, contrasting it with Bitcoin's undefined purpose.
Pushes back on Bitcoin as digital gold, arguing no successful banking or credit system has ever run on gold because of its volatility.
Asks why Bitcoin still lacks obvious low-risk upgrades like Schnorr signatures and NiPoPoWs years after they were proposed, dismissing the idea that decentralization alone will push it toward a coherent goal.
Argues that Bitcoin's decentralization excuse lets the community dodge real coordination or focus, and that they mostly just measure success by price rather than actual adoption.
Banters about preferring Goodfellas over The Godfather and loving Joe Pesci and Casino.
Recalls the old, egoless Bitcoin community from the 2013 San Jose conference and contrast it with the commercialized, diamond-watch Miami Bitcoin conference.
Remembers the 2014 Miami Bitcoin Help Center booth where core developers freely answered anyone's questions, unlike today's culture of superstar lines.
Traces his own start as a free Bitcoin educator on Udemy answering thousands of student emails, and recall how Mike Hearn was pushed out of Bitcoin over the block size wars despite once emailing Satoshi directly.
Praises Alex Chepurnoy's Ergo as the true spiritual successor to Bitcoin, keeping proof of work and UTXO but adding smart contracts and a zero ego community.
Ergo became a beta testbed for Cardano's extended UTXO smart contracts, much like Microsoft prototyping the Xbox 360 on Apple hardware.
Credits Alex's earlier Scorex framework, a modular pedagogical blockchain toolkit, for spawning both Ergo and other chains like the Waves protocol.
Highlights the Sigma protocol as a way of expressing scripts as concise proofs that can be checked correct without rerunning the whole program, a promising idea from a researcher at Boston University.
Points to Microsoft Research's Pinocchio and Geppetto papers on verified computing, generating a proof that a computation was done correctly instead of redoing the work, tying it back to P versus NP where checking is often far cheaper than solving.
Read from the Extended UTXO Model paper explaining why Bitcoin's UTXO and Ethereum's account model diverge in expressiveness.
Compares UTXO to cash register accounting, where you hand over a whole 20 dollar bill and get atomic units of change back rather than tearing bills apart.
Bitcoin's original UTXO model had no way to attach complex conditions to a payment, like only releasing funds once a specific event happens, unlike a mutable bank ledger.
Ethereum's account model as bank style mutable ledger accounting, versus UTXO where each script's state stays local instead of needing a global view.
Details the Homomorphic Ledgers paper showing UTXO, extended UTXO and account models are isomorphic, so properties map between the two systems.
Emphasize that Plutus contracts are deterministic, what you test locally is exactly what runs on chain, unlike Ethereum where mutable global state can misprice and fail a transaction.
Admits extended UTXO is a bit less expressive at the edges but that it keeps Bitcoin style developers in familiar territory and maps naturally onto functional programming.
Reassures that building on eUTXO in practice isn't too different, developers can still build Uniswap style dApps, stablecoins and oracles with some new patterns, as shown in the Plutus Pioneers lectures.
Jokes that building smart contracts without oracles is like trying to have sex with your pants on, since contracts need outside world data to be useful.
Praises Wolfram Alpha as a computable, endlessly cross-queryable data source built from Stephen Wolfram's obsessive data hoarding, ideal as an oracle feed.
Flags that Wolfram's data is centralized, while Chainlink represents a more academic, decentralized philosophy for aggregation, injection and veracity incentives.
Cornell's Town Crier project, which used SGX trusted hardware to scrape data or run contract computation off chain without bias or tampering.
Uses the Simpsons Mr. Burns back door bit to illustrate that a smart contract system is only as secure as its weakest oracle data feed.
Proposes that stake pools could double as decentralized oracle or data feed providers once the system admits heterogeneity, with no need for a separate oracle token since you can just use ADA.
A vision of a distributed Wolfram Alpha feed acting like a Bloomberg terminal for Cardano, positioned as an open alternative alongside Chainlink rather than a competitor to kill.
The oracle design spectrum runs from fully decentralized aggregation to fully centralized feeds, and developers should get to choose, citing World of Warcraft as an inherently centralized case.
Contrasts Blizzard's single company control of game item minting with an algorithmic stablecoin's need to aggregate many exchange prices, where a bad feed mispriced the coin and everything goes to hell.
Sums up oracle design as building standardized interfaces so developers can pick their own injection method and assurance level, calling Wolfram's data set second to none.
Confirms it's possible to build games where item ownership sits outside Blizzard's control, like a decentralized version of the Diablo grandfather sword.
Points to CryptoKitties as proof that game assets can live on a blockchain ledger independent of the game's own servers, so anyone can revive the game later.
Argues blockchain ownership fixes gaming economics by letting an in-game economy generate a long revenue tail like Eve Online or Second Life, instead of front-loading 95 percent of value in the first six months.
Speculates that as AI and automation raise living standards, people will find meaning by playing video games and living out ambitions like being a virtual Bill Gates.
Predicts most people will prefer virtual reality once photorealism crosses the uncanny valley, since actual reality is hard.
Warns of a dark future industry engineering amnesia so people forget they're inside a virtual world, then muse that love and beauty are themselves a kind of forgetting the mess of life and death.
Considers it entirely possible humanity already lives in a simulation, arguing it wouldn't matter as long as society doesn't collapse from a mass retreat into virtual worlds.
VR and game engines are evolving fast, evidenced by billions of hours spent in games like Skyrim, and frame smart contracts as the mechanism that migrates meatspace value into that digital world.
Names Alonzo after mathematician Alonzo Church, whose lambda calculus underlies Cardano's functional approach, with smart contracts set to arrive via an invisible hard fork combinator event.
Walks through the Alonzo rollout starting from the current Mary era's multi-asset support, through Alonzo Blue's successful testnet fork proving Mary to Alonzo transactions and stake pool operator readiness.
Outlines the remaining Alonzo White and Purple testnet stages, with Purple as an open devnet for every developer to test Plutus contracts before the public hard fork, like sending Xbox dev kits ahead of a console launch.
Stresses that once shipped to over a hundred exchanges and countless wallets, evolving Cardano becomes less a technical problem and more a coordination problem across every constituency.
Jokes about sleeping like a baby waking up crying every two hours, then admit what really worries him is being judged as much by the apps built on Cardano as by the platform itself, citing the DAO hack unfairly blamed on Vitalik.
Wants a green checkmark style assurance system so users know a deployed smart contract was audited or followed a spec, and reveals a smart contract engineering institute being set up with the University of Wyoming.
DEXs aim to remove custodial risk from exchanges, citing Mount Gox's 2013 collapse and the danger of a hacked or regulator shut down centralized exchange.
Acknowledges DEXs face latency disadvantages against high frequency centralized trading, but credits protocols like Uniswap for evolving to offset those security tradeoffs with liquidity and openness.
Sees decentralized exchanges as replacing the old centralized exchange gatekeepers, recalling how Bitfinex once asked for $5 million just to list ADA.
Calls regulation, not the technology, the biggest existential problem for DEXes since they take in value with no KYC or AML layer.
Proposes regulated DEXs built around decentralized identifiers as a way to add compliance without a single centralized gatekeeper.
Recalls making an explainer video for Jack Dorsey after his Bitcoin changes everything comment, framing proof of work and proof of stake as just different resource trade offs, not a security hierarchy.
Argues the real question is how to evolve from proof of work to a richer end state, pointing to Permacoin's proof of storage as a way to fund storing IPFS style data.
Envisions balancing proof of stake, proof of storage, and proof of useful computation together to build a system that's simultaneously the world's largest programmable computer and database.
Jack Dorsey that adding Cardano to Square or Cash App should be a market decision, not tribal loyalty, noting Coinbase and Erik Voorhees both moved past Bitcoin maximalism for business reasons.
Challenges proof of work maximalists to actually define what makes their system more secure, comparing blind faith in Satoshi's design to a religion that never offers real evidence.
Makes the case for Cardano to Elon Musk and Tesla on sustainability grounds, arguing proof of stake is built to minimize energy use while proof of work's efficiency gains just fuel a bigger ASIC arms race.
Argues Bitcoin is the least programmable cryptocurrency, so Tesla style ideas like cars sharing data over a 5G or IoT network would need Ethereum or Cardano style infrastructure instead.
Dogecoin's origin as a joke, a copy of a copy running from Bitcoin to Litecoin to Dogecoin, that unexpectedly grew into a huge community despite starting as a parody.
The Elon directed Doge video was meant to show what a real third generation cryptocurrency overhaul would require, pointing to Solana, Harmony, and Cardano as examples already doing that work.
Warns that Dogecoin's ownership is so concentrated, likely under 100 wallets holding the bulk of supply, that early holders could dump on retail at any price and still profit.
Shares a story about a Longmont waitress who bought Dogecoin only because she saw Elon tweet about it, worried that uninformed retail buyers are the ones who get hurt.
Hopes real developers step in to give Dogecoin genuine utility and a value floor, staying open to helping clean up its aging codebase alongside Cardano.
Admires Elon's combination of humor and extreme ambition in the face of impossible odds, calling that same bold spirit the appeal behind Dogecoin.
Questions why Elon doesn't put that same engineering ambition behind one of the many serious crypto platforms actually solving real problems instead of Dogecoin.
Weighs whether crypto lacks the first principles risk taking that drove Tesla and SpaceX, contrasting Bitcoin's hunger for bold ideas with Cardano's push for more rigor and formalism.
Muses that the most entertaining outcome might be the most likely one, tying the viral power of fun and dopamine to why something like Dogecoin takes off.
The personal cost of cleaning up after Elon's market moving tweets, having to face a Senate inquiry while Musk laughs about it with friends, but stays hopeful he'll eventually build something real in crypto.
Get asked what Hydra is and how it stacks up against rollups, explaining that all layer two designs trade some base layer security and decentralization for fast, cheap settlement.
Hydra's novelty over Lightning is handling rich off chain state, like a DEX or a game, rather than just simple point to point payments.
Outlines Hydra's roadmap from a single head design to multi head composition with a routing protocol and tail protocols so funds survive nodes going offline.
Contrasts Hydra's interactive channels with rollups and recursive SNARKs, like the Mina protocol, which instead batch transactions into one small, logarithmically verifiable proof.
Layer two channel networks could eventually bridge Hydra and Lightning together into a cross chain interoperability network using wrapped assets.
Compares layer two protocols to biological cell differentiation, each one adding a specialized capability like a limb or a sense rather than competing to be the single winner.
Explains why identity should live at layer two instead of Cardano's base layer, letting users bring it in only when they need things like human readable addresses.
Get asked about non-interactive proofs of proof of work, explaining how rare high difficulty blocks let you build a small compressed proof for a long range of chain history.
NIPoPoWs enable light wallets and side chains to verify history without holding the full blockchain, crediting the idea to Dionysis Zindros and Aggelos Kiayias without requiring any hard fork.
Exploring NIPoPoWs for sharded proof of work systems like Prism, plus a related paper on log space mining that lets miners work from a tiny compressed ledger.
Shares Dionysis Zindros's story, a PhD student under Aggelos now doing a postdoc under David Chaum at Stanford, whose 2016 NIPoPoWs paper took three or four years to survive peer review.
Takes on the top Reddit AMA question and play devil's advocate on how Cardano could fail, starting with the project never achieving genuine self-evolution and decentralization.
A dystopian failure mode where the system self-evolves toward centralized control, comparing it to China's social credit system.
Raise losing network effect through user churn as another failure mode, but note Cardano's tattoo level fan loyalty makes that unlikely.
Hopes the Cardano community stays open and Socratic rather than repeating Bitcoin's slide into toxic, dogmatic maximalism.
Feel like a failure whenever people only celebrate ADA's price instead of the deeper mission to re-engineer voting, property, and payments around user control.
Acknowledges risks like losing project cohesion, but argue the open, patent free research and code being permanently in the commons means the mission survives even if the company disappeared.
Flags bugs as a real failure risk, pointing to Zcash's shielded pool flaw as an example that can permanently damage confidence in a coin's monetary policy.
Argues Cardano's heavy use of formal methods and peer review makes a catastrophic hidden bug less likely than in other protocols, since the process has already caught weird bugs before production.
Polkadot as a close competitor, describing plans to fork and run private versions of its code to study it, the same way ConsenSys tore down EOS.
Polkadot's consensus borrows heavily from Ouroboros since Gavin Wood studied Cardano's research before designing it.
Contrasts Polkadot's disciplined, commercially predictable rollout with Ethereum's spray and pray approach to growing its ecosystem.
Lists Cardano's technical edges over Polkadot: native multi-asset support, a more expressive ledger, and faster proof of stake evolution through Ouroboros Omega and Mithril.
Argues Cardano thought harder about governance than Polkadot, building Catalyst and Voltaire so every ADA holder can participate instead of leaving governance for later.
Calls Polkadot Ethereum 1.5, telling the alternate history where Ethereum could have become IBM's enterprise blockchain or adopted the Snow White protocol had different licensing and roadmap choices been made.
Thanks Lex for the personal video made before the conversation, saying it came from wanting an honest talk about the technology and vision rather than clickbait journalism.
Compares being asked about the six months at Ethereum to asking Paul McCartney about John Lennon, joking about a future reunion interview with Vitalik moderated by an AI version of Walt Mossberg.
Reflects fondly on years of building Cardano, traveling to 52 countries, meeting heads of state, and helping pass 24 crypto friendly laws in Wyoming.
Shares plans for an AGI research project with Ben Goertzel based in Rwanda, joking about Goertzel's ever present hat.
Admits frustration at still being called an Ethereum co-founder in every press story despite barely speaking to Vitalik anymore, and warn that the internet lets fabricated stories spread as memes that half the world ends up believing.
Enjoying both the physical and digital worlds while creating, then reflect that most people, including himself, will ultimately be forgotten in the long arc of history like a single line from Hitchhiker's Guide to the Galaxy.
Jokes around with Lex about UFOs talking to dolphins, guessing the sightings are actually hypersonic aircraft being tested over the ocean.
Answers whether power and money corrupt, arguing it is most dangerous for people who inherit wealth or a title without being prepared for it, citing dynasties like the Mellons and Rockefellers.
Uses owning a 400 head bison ranch in Wyoming as an example of how unrestrained crypto wealth lets you do whatever you want with no one telling you no.
Recommends staying grounded through animals and gardening, since that kind of honest physical work does not care whether you are a billionaire or broke.
Credits giving things away, including 293,000 ether worth over 1.2 billion dollars at its peak that he gave to his secretary, with bringing back even more in return.
Cites George Washington staying connected to Mount Vernon during the war as proof leaders should stay grounded, and argue that a mindset of impermanence, living for the next generation, keeps you gracious and collaborative.
Criticizes Xi Jinping for dismantling the checks and balances China built after Mao to prevent another cult of personality.
Compares unchecked power to Rome after Augustus, when successors like Caligula and Nero destroyed the republic's legacy, and recall being prepared to lose everything, as he did for two years after leaving Ethereum with a damaged reputation and no one calling.
Credits surrounding himself with great people, instead of being narcissistic and transactional, with rebuilding after Ethereum, tracing that instinct to his father and Montana homesteader ancestors who survived by taking care of their neighbors.
Announces putting 20 million dollars into a center for automated theorem proving that heavily invests in the Lean language to give back to the mathematics community.
The interesting tension that Bitcoin avoids leadership corruption entirely because Satoshi Nakamoto remains anonymous and the project has no leader at all.
Argues the best leaders are felt but never seen, since staying out of the room inspires new leaders instead of freezing a company the way reverence for Walt Disney and Steve Jobs did after they died.
Jokes about being a poker playing robot before naming Adam Back as the most likely Satoshi Nakamoto by Occam's razor, pointing to Bitcoin's Forth style scripting quirks and his invention of Hashcash.
A US Army code stylometry technique that could fingerprint Satoshi Nakamoto's coding style against old open source repositories with about 94 percent accuracy.
Natural language stylometry analysis of Satoshi Nakamoto's writings points closest to Nick Szabo, though he has not personally verified the methodology.
Publishing anonymously hurts academics because tenure, co-authorship credit, and future deals all depend on your work being publicly attributed to you.
Highlights Q Editus, a project inherited from an anonymous pseudonymous mathematician named Bill White, that builds a marketplace paying people to write mathematical proofs in a theorem prover.
Lays out a vision for decentralized governance, arguing legitimacy sits on a spectrum between an outright dictator and a pure democracy where nothing ever gets done.
Contrasts Bitcoin and Ethereum's implicit, informal improvement proposal process with Cardano's explicit governance built around the Cardano Improvement Proposal system and formal e-voting research.
Cardano's treasury, now holding over a billion dollars of ADA from protocol inflation, is controlled by the community through Catalyst rather than by himself or the foundation.
A plan to graduate the community from voting on simple protocol parameters up to voting on hard forks once the social dynamics and a constitution are in place.
Details the partnership with IdeaScale that hosts Catalyst proposal discussions, plus a new incubator and accelerator being built by chief of staff Tamara Hassan.
Funded Catalyst proposals need an accountable auditor as a real world counterparty, since a blockchain itself cannot hold anyone responsible.
Acknowledges that writing a constitution is hard since natural language invites multiple interpretations like poetry, and points to formal languages like Idris, Coq, and Agda as ways to eventually reduce that ambiguity.
Walks through CIP 007, a real Cardano Improvement Proposal that simplifies the stake pool pledge formula so smaller pool operators get a better deal.
Argues a constitution must define voting thresholds up front, citing Brexit as an example of how a single up or down vote produced buyer's remorse that a multi stage vote might have avoided.
Champion preference and ranked choice voting systems like Condorcet and Borda for avoiding wasted votes, while acknowledging Arrow's paradox means no voting system is perfect.
Fully decentralizing the protocol's research brain by letting academic peer review and tenure incentives drive an independently funded, growing pile of over 100 papers about Cardano.
Sets a goal of having at least three independent client teams and codebases for Cardano, separating commercial clients from a formally specified reference blueprint.
Argues the governance tools being built for Cardano are completely reusable for nation states, companies, and any dApp, pointing to Ethiopia's digital identity rollout as an early example.
Recaps four years spent in Ethiopia training developers and building a relationship with the Ministry of Education, admitting projects there always take longer than the six months first promised.
The Atala PRISM based decentralized identity system giving every Ethiopian student a DID to prove credentials like GPA and diplomas, calling it the largest blockchain deal of its kind and expecting it to reach 20 million people.
Reveals that Ethiopia's Prime Minister is himself a cryptographer who can read the team's papers and has written a Digital Ethiopia 2025 agenda targeting national digital identity and e-voting.
Floats rolling out preference voting and online voter registration for Wyoming's party primaries in 2022, framing distrust in the 2020 US election as an opening for e-voting reform.
Weighs in on El Salvador making Bitcoin legal tender, arguing reciprocal currency recognition between nations could let it trade on forex exchanges as a clever backdoor around regulators.
Recalls a half joking old plan to get a small nation like Tuvalu to adopt crypto, and mention knowing El Salvador's president's brother and possibly visiting.
Distinguishes political optics from real commitment, warning that autocratic leaders who adopt blockchain do not realize the technology inevitably pushes power away from them.
China's Bitcoin crackdown as a reaction once regulators realized it threatened capital controls and their own digital yuan ambitions, after initially tolerating it for elite corruption.
Predicts a ripple effect where more nations adopting crypto as legal tender pressures superpowers like the US, China, and Russia and pushes cryptocurrency toward becoming a global store of value.
Breaks transactions down into five programmable properties: the asset, the transfer pattern, metadata, the smart contract relationship, and the regulatory jurisdiction.
Programmable compliance as pulling in a ready made regulatory library, like one for air travel, instead of hiring lawyers to translate rules into a contract by hand.
Calls the crypto industry's efficiency gains and universal asset representation, his financial stem cell idea, an inevitable victory complicated only by superpower rivalry between the US and China over who sets the global standard.
Agrees wholeheartedly that pushing power to the edges through crypto can alleviate suffering caused by centralized power, corruption, and abuse.
Frames the crypto industry as the antidote to predatory finance, pointing to 85 percent interest loans to the poor, costly remittances, and money laundering by unpunished elites.
State the industry's only real goal is fairness, not guaranteed wins, and tie Bitcoin's creation directly to the 2008 financial crisis.
Points to Wyoming's fully reserved SPDI banks as proof of banking hypocrisy, since fractional reserve banks fear default from institutions that actually hold a dollar for every dollar.
Expresses deep passion for Africa's young population who are ready to discard broken legacy systems, believing 1.2 billion people could leapfrog into a fairer system within 20 years.
Argues capitalism as an engine is fine, it is just mis-parameterized for the short term, and crypto tokenomics can pay people today for benefits felt generations later.
Wants society judged by how its least well off members are doing, not GDP, with crypto acting as a vanguard that pressures other nations to change through envy and ripple effects.
Always feel the burden of leadership, since employees and their families depend on him and he is never allowed to have a bad day.
Learned to delegate and trust his team even when they fail, and refuses to ever file a patent so the work belongs to whoever carries it forward after him.
Points to Norman Borlaug as proof one person's knowledge transfer can save a billion lives even without becoming a household name.
Expected the most likely outcome of building crypto was prison, since financial regulations were never built for rapid monetary innovation like Bitcoin.
Gets asked what a productive day in his life actually looks like, from diet to work schedule to social media.
Alternates a dry year and a wet year with alcohol so he never overdoes any one habit, joking that 2020 was his dry year.
Practices 16:8 intermittent fasting and once pushed an extended fast all the way to two full weeks.
Recounts fainting near the two week mark of that fast and calling his doctor brother to learn how to safely start eating again.
Besides fasting, leans on meditation and a red light photobiomodulation headset he believes supports neurogenesis and blood flow.
Misses the long uninterrupted deep work sessions he had before becoming CEO and is trying to carve out no-meeting Fridays.
Uses float tanks and daily naps to clear stress and sharpen productivity and clarity.
Genuinely enjoys the livestreams and calls the community a cult following of tens of thousands who recognize him everywhere he travels.
Finds mushrooms fascinating for both personal and business reasons, citing zombie cordyceps fungus as one of nature's strangest tricks.
Highlights lion's mane and other medicinal mushrooms showing promise against Alzheimer's, depression, Parkinson's, and even COVID.
Cites Johns Hopkins research showing psilocybin performed as well as or better than SSRIs for treating depression.
Believes psychedelics could let terminal cancer patients find peace and dignity in their final months.
Backs Farmbox Foods, a venture growing hundreds of pounds of gourmet mushrooms a week in shipping container farms with strong margins.
Loves the mushroom hunting community, recalling a beetle and bolete researcher he met at a festival, and notes how many species remain undiscovered.
Pushes for local aquaponics and hydroponics facilities near every major city after seeing barren grocery shelves during COVID.
Reflects on fans comparing his and Joe Rogan's audiences before explaining how he would approach a conversation with Rogan.
The way to explain crypto to a newcomer like Joe Rogan is to start from something they already love, such as elk tags, not the technology itself.
NFTs enabling perpetual royalties baked into a smart contract that outlive the creator, citing Warhol or Picasso as examples.
Proof of work versus proof of stake once someone is hooked, the same curiosity that pulled him into rereading 20th century monetary history and how money can control populations.
Finds Cardano's governance and voting reform far more exciting than the number go up financial conversation that dominates crypto discourse.
Trades banter over favorite games like Diablo and Skyrim, leading into a deep dive on Elder Scrolls history.
Reveals he bought the intellectual property for Legends of Valor, the obscure unfinished game that originally inspired The Elder Scrolls.
Daggerfall's randomly generated massive world made it feel like a place you could live in forever.
Plans to build Legends of Ellara as a modernized overhaul of Legends of Valor once he finally has time for the passion project.
Wants to bring algorithmically generated music and AI-driven dialogue like the game Event Zero into future game design.
Worries Microsoft's acquisition of Bethesda and its employee exodus mean Elder Scrolls 6 will disappoint, and jokes his Legends of Valor remake could become a spiritual successor.
The Haskell framework Euterpe for algorithmic music and credits Jeremy Soule's Baldur's Gate score as proof of how much a soundtrack shapes a game's lasting feeling.
Ranks Baldur's Gate and Arcanum among the greatest games ever, describing Arcanum's world where magic and technology destroy each other.
Recalls talking Arcanum's final villain into suicide through pure persuasion, and praises Planescape Torment's amnesiac Nameless One story.
Believes old story-driven games still hold up because video games remain one of the last mediums that commands people's full attention.
Advises young people to learn how to learn, value emotional intelligence as much as raw IQ, and mentor others early on.
Warns against conflating your passion with your paycheck, since forcing them together tends to cause burnout.
Gets asked for the meaning of life and jokingly deflects first about the king of the rats comment.
The story of a Japanese shogun who wandered thirty years for the perfect cherry blossom and died beside one, arguing the journey itself is what matters.
Compares meaning to Sisyphus finding purpose in rolling his stone and Jiro spending seventy years perfecting sushi, joy found in the act itself.
Since human beings are just a blip against the size of the universe, the point of life is finding happiness regardless of circumstance and sharing that mindset with others.
Gets asked what his own Old Man and the Sea style struggle is, the big fish he will wrestle with at the end of his life.
Still loves mathematics and dreams of proving the Goldbach conjecture, now leaning on automated theorem proving as he gets older.
Another white whale is helping George Church raise cloned woolly mammoths, which he hopes to host on his Wyoming ranch among the bison.
Jokes about wanting to catch and recognize the same giant catfish decades apart as a marker of a life well lived.
Thanks Lex for the conversation, calling it an honor and reciprocating how much fun the exchange was.
Closes the episode by thanking the sponsors and reading a William Faulkner quote about losing sight of the shore to find new horizons.