Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

First Principles Ep. 3 with Leslie Lamport Before blockchains could reach consensus, Leslie Lamport had to define what agreement even meant when computers fail, lie, or disappear. Hosted by Tim Roughgarden with Ittai Abraham. 00:00 The problem every blockchain is built to solve 02:52 Why concurrent systems are surprisingly...

a16zcrypto's profile picture

a16z crypto

1,119,691 subscribers

65,334 Aufrufe • vor 1 Monat •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

First Principles Ep. 1 Bitcoin gets credited with inventing trustless consensus. It didn’t. The problem was named in the 1980s, and it took the research world years to even notice Satoshi had solved it. A new series hosted by Tim Roughgarden, with Ittai Abraham. 00:00 Introduction to First Principles: The Scientific Roots of Blockchain Technology 00:56 Why consensus matters for blockchains 02:30 Byzantine agreement: The old computer science problem Bitcoin made practical 04:34 Blockchains as a shared system of record: State machine replication and blockchain state 06:41 How two research worlds — distributed computing and crypto — began to converge 07:49 Proof of work vs. proof of stake 09:27 Why Ethereum’s move to proof-of-stake took years 11:08 When crypto rediscovered decades of distributed systems research 12:49 Throughput, latency, and modern consensus design 14:05 DAG-based protocols and faster blockchains 15:25 Peace time vs. war time: why modern blockchains need two modes 16:47 Theory, practice, and the future of blockchain research
20:00

First Principles Ep. 1 Bitcoin gets credited with inventing trustless consensus. It didn’t. The problem was named in the 1980s, and it took the research world years to even notice Satoshi had solved it. A new series hosted by Tim Roughgarden, with Ittai Abraham. 00:00 Introduction to First Principles: The Scientific Roots of Blockchain Technology 00:56 Why consensus matters for blockchains 02:30 Byzantine agreement: The old computer science problem Bitcoin made practical 04:34 Blockchains as a shared system of record: State machine replication and blockchain state 06:41 How two research worlds — distributed computing and crypto — began to converge 07:49 Proof of work vs. proof of stake 09:27 Why Ethereum’s move to proof-of-stake took years 11:08 When crypto rediscovered decades of distributed systems research 12:49 Throughput, latency, and modern consensus design 14:05 DAG-based protocols and faster blockchains 15:25 Peace time vs. war time: why modern blockchains need two modes 16:47 Theory, practice, and the future of blockchain research

a16z crypto

97,075 Aufrufe • vor 1 Monat

First Principles Ep. 2 w/ Barbara Liskov Every blockchain alive today leans on replication ideas worked out in the 1980s, by a Turing Award winner who wasn’t thinking about how it might apply to money at all. Hosted by Tim Roughgarden with Ittai Abraham. 00:00 How do systems stay reliable when parts fail? 01:18 Barbara Liskov’s path from programming languages to distributed systems 05:45 Why modularity is “everything” 07:22 The replication problem: keeping data available across many machines 09:58 Viewstamped replication and the “ledger” before blockchains 16:32 Why good research starts with what you don’t understand 18:10 Leslie Lamport, Paxos, and the inevitability of ideas in the right time, in the right place 21:48 Practical Byzantine Fault Tolerance: what changes when replicas can lie 19:35 How PBFT bridged theory and practical systems 22:38 Why you should never trust an individual replica 28:39 Why blockchains are state machine replication in the wild 31:27 AI, verification, and the future of computer science
35:46

First Principles Ep. 2 w/ Barbara Liskov Every blockchain alive today leans on replication ideas worked out in the 1980s, by a Turing Award winner who wasn’t thinking about how it might apply to money at all. Hosted by Tim Roughgarden with Ittai Abraham. 00:00 How do systems stay reliable when parts fail? 01:18 Barbara Liskov’s path from programming languages to distributed systems 05:45 Why modularity is “everything” 07:22 The replication problem: keeping data available across many machines 09:58 Viewstamped replication and the “ledger” before blockchains 16:32 Why good research starts with what you don’t understand 18:10 Leslie Lamport, Paxos, and the inevitability of ideas in the right time, in the right place 21:48 Practical Byzantine Fault Tolerance: what changes when replicas can lie 19:35 How PBFT bridged theory and practical systems 22:38 Why you should never trust an individual replica 28:39 Why blockchains are state machine replication in the wild 31:27 AI, verification, and the future of computer science

a16z crypto

58,704 Aufrufe • vor 1 Monat

Leslie Lamport won a Turing award for his fundamental contributions to distributed systems. For instance, he invented the Paxos consensus algorithm that is a critical component of many distributed systems today. I interviewed him about his work and career. We discussed: • Why he never considered himself smart • The stories behind Paxos and Byzantine Generals Problem • Experiences working with Dijkstra • Paxos vs Raft Algorithms • How to improve your thinking Where to watch: • YouTube: • Spotify: • Apple Podcasts: • Transcript:
1:17

Leslie Lamport won a Turing award for his fundamental contributions to distributed systems. For instance, he invented the Paxos consensus algorithm that is a critical component of many distributed systems today. I interviewed him about his work and career. We discussed: • Why he never considered himself smart • The stories behind Paxos and Byzantine Generals Problem • Experiences working with Dijkstra • Paxos vs Raft Algorithms • How to improve your thinking Where to watch: • YouTube: • Spotify: • Apple Podcasts: • Transcript:

Ryan Peterman

338,818 Aufrufe • vor 5 Monaten

ARC-AGI is redefining how to measure progress on the path to AGI - focusing on reasoning, generalization, and adaptability instead of memorization or scale. At NeurIPS 2025, YC's Diana sat down with ARC Prize President Greg Kamradt to find out why most AI benchmarks fail, how ARC-AGI reveals the limits of today’s models, and why measuring intelligence may be harder than building it. 00:11 — What ARC Prize is and why it exists 00:38 — François Chollet’s definition of AGI 01:48 — What ARC-AGI Actually Tests 02:25 — When LLMs Failed the ARC Benchmark 03:38 — ARC-AGI Becomes the Standard 04:49 — False Positives in AI Progress 06:06 — The Evolution of ARC-AGI 08:55 — Measuring Intelligence beyond just accuracy 10:25 — What happens if a model solves ARC-AGI?
11:59

ARC-AGI is redefining how to measure progress on the path to AGI - focusing on reasoning, generalization, and adaptability instead of memorization or scale. At NeurIPS 2025, YC's Diana sat down with ARC Prize President Greg Kamradt to find out why most AI benchmarks fail, how ARC-AGI reveals the limits of today’s models, and why measuring intelligence may be harder than building it. 00:11 — What ARC Prize is and why it exists 00:38 — François Chollet’s definition of AGI 01:48 — What ARC-AGI Actually Tests 02:25 — When LLMs Failed the ARC Benchmark 03:38 — ARC-AGI Becomes the Standard 04:49 — False Positives in AI Progress 06:06 — The Evolution of ARC-AGI 08:55 — Measuring Intelligence beyond just accuracy 10:25 — What happens if a model solves ARC-AGI?

Y Combinator

98,369 Aufrufe • vor 8 Monaten

Enough of the fluff. Time for some real advice. Launch your GPT 5.6 Sol Pro. Use the following prompt (feel free to swap the second and third lines for something of your choice): Prove for me that the group of rational points on the elliptic curve (y^2=x^3+17) is finitely generated. I want to tailor the general proof to this particular case to narrow down my understanding to a particular case. Write down a full mathematical proof, step by step, in Leslie Lamport style. Unpack all the definitions and notions so that it is fully self-contained. An example here: Notice how Sol is spawning all the subagents to complete the task. What is "Leslie Lamport style"? It's a technique where you split your proof into short, one-line sentences (claims) that are easy to justify. In a situation where your proof is not a 1-to-1 translation of, e.g., Python code, the Leslie Lamport-style proof can be treated as such. It tremendously improves the possibility of verifying each claim and helps the model structure the arguments in a very clear way. Leslie Lamport, a legendary mathematician and computer scientist, was a promoter of this way of structuring proofs. Partly because of that, he created an extension of Donald Knuth’s typesetting system TeX called LaTeX. Here you can read about "Leslie Lamport-style proofs":
1:44

Enough of the fluff. Time for some real advice. Launch your GPT 5.6 Sol Pro. Use the following prompt (feel free to swap the second and third lines for something of your choice): Prove for me that the group of rational points on the elliptic curve (y^2=x^3+17) is finitely generated. I want to tailor the general proof to this particular case to narrow down my understanding to a particular case. Write down a full mathematical proof, step by step, in Leslie Lamport style. Unpack all the definitions and notions so that it is fully self-contained. An example here: Notice how Sol is spawning all the subagents to complete the task. What is "Leslie Lamport style"? It's a technique where you split your proof into short, one-line sentences (claims) that are easy to justify. In a situation where your proof is not a 1-to-1 translation of, e.g., Python code, the Leslie Lamport-style proof can be treated as such. It tremendously improves the possibility of verifying each claim and helps the model structure the arguments in a very clear way. Leslie Lamport, a legendary mathematician and computer scientist, was a promoter of this way of structuring proofs. Partly because of that, he created an extension of Donald Knuth’s typesetting system TeX called LaTeX. Here you can read about "Leslie Lamport-style proofs":

Bartosz Naskręcki

31,115 Aufrufe • vor 1 Monat

First Principles Ep. 5 with Paul Milgrom One of the most important auction designs in modern history was sold to the FCC on a 3.5-inch disk. Nobel Prize winner Paul Milgrom on auctions, price discovery, and how mechanism design became real-world infrastructure — from spectrum auctions to DeFi. Hosted by Tim Roughgarden with Scott Kominers. 00:00 Intro: economics assumptions that are “just wrong” 02:19 Scott Kominers and Tim Roughgarden on the genius of Paul Milgrom 05:35 An intro to the Glosten-Milgrom model: The paper that created entire fields of economics 07:48 The auction theory breakthroughs of the 1980s 17:15 Why market microstructure matters for DeFi 24:17 When math teaches economics something new 32:38 How theory became spectrum auction design 36:22 The floppy disk that helped convince the FCC 41:05 What changed when auctions moved online 45:28 The auction that reorganized television 1:07:30 What economics and computer science can learn from each other 1:13:22 Futures markets for compute 1:15:12 Paul Milgrom’s advice for builders
1:18:00

First Principles Ep. 5 with Paul Milgrom One of the most important auction designs in modern history was sold to the FCC on a 3.5-inch disk. Nobel Prize winner Paul Milgrom on auctions, price discovery, and how mechanism design became real-world infrastructure — from spectrum auctions to DeFi. Hosted by Tim Roughgarden with Scott Kominers. 00:00 Intro: economics assumptions that are “just wrong” 02:19 Scott Kominers and Tim Roughgarden on the genius of Paul Milgrom 05:35 An intro to the Glosten-Milgrom model: The paper that created entire fields of economics 07:48 The auction theory breakthroughs of the 1980s 17:15 Why market microstructure matters for DeFi 24:17 When math teaches economics something new 32:38 How theory became spectrum auction design 36:22 The floppy disk that helped convince the FCC 41:05 What changed when auctions moved online 45:28 The auction that reorganized television 1:07:30 What economics and computer science can learn from each other 1:13:22 Futures markets for compute 1:15:12 Paul Milgrom’s advice for builders

a16z crypto

33,605 Aufrufe • vor 29 Tagen

It started on a napkin. Turing Award winner Leslie Lamport on sketching a piece of modern cryptography in a Berkeley coffee shop, and how the fault model behind every blockchain came from research on keeping airplanes in the sky: "I was one of the few people in the world who happened to have heard about digital signatures… So I just wrote, literally on a napkin, a simple algorithm for writing digital signatures." "They had to assume that it could do anything, and therefore they were faced with the Byzantine generals problem… Once hackers started appearing on the scene, it was obvious that you really did need Byzantine fault tolerance."
1:55

It started on a napkin. Turing Award winner Leslie Lamport on sketching a piece of modern cryptography in a Berkeley coffee shop, and how the fault model behind every blockchain came from research on keeping airplanes in the sky: "I was one of the few people in the world who happened to have heard about digital signatures… So I just wrote, literally on a napkin, a simple algorithm for writing digital signatures." "They had to assume that it could do anything, and therefore they were faced with the Byzantine generals problem… Once hackers started appearing on the scene, it was obvious that you really did need Byzantine fault tolerance."

a16z crypto

53,101 Aufrufe • vor 1 Monat

First Principles Ep. 6 with Ron Rivest Long before the internet became a place to bank, transact, or build blockchains, Turing Award winner Ron Rivest helped solve a foundational problem: how can strangers communicate securely without first meeting to exchange a secret? Rivest tells the story of inventing RSA with Adi Shamir and Leonard Adleman, why digital signatures fascinated him even more than encryption, and how cryptographic hashes, government pressure, patents, and standards shaped the security infrastructure we rely on today. Rivest shares his unusually candid views on quantum computing, post-quantum security, and more. Hosted by Tim Roughgarden with Dan Boneh. 00:00 Intro: the two problems that could break modern cryptography 01:10 Why Ron Rivest’s work underpins the internet and blockchains 08:10 Before public-key cryptography, there was no theory of security 11:05 The open problem that led to RSA 13:55 The night Ron Rivest discovered the core idea behind RSA 17:55 Why digital signatures were the real breakthrough 19:14 The RSA challenge — and a prediction that was off by quadrillions of years 28:33 How government pressure shaped cryptographic standards 30:36 Designing the hash functions that made digital signatures practical 33:26 The Fiat–Shamir transformation, explained 38:04 Building a cryptography company before the web existed 42:31 Will quantum computers ever become powerful enough to break RSA? 48:02 The cryptography securing the internet, elections, and everyday life 50:11 What surprised Dan: Quantum giveth and quantum taketh away
54:44

First Principles Ep. 6 with Ron Rivest Long before the internet became a place to bank, transact, or build blockchains, Turing Award winner Ron Rivest helped solve a foundational problem: how can strangers communicate securely without first meeting to exchange a secret? Rivest tells the story of inventing RSA with Adi Shamir and Leonard Adleman, why digital signatures fascinated him even more than encryption, and how cryptographic hashes, government pressure, patents, and standards shaped the security infrastructure we rely on today. Rivest shares his unusually candid views on quantum computing, post-quantum security, and more. Hosted by Tim Roughgarden with Dan Boneh. 00:00 Intro: the two problems that could break modern cryptography 01:10 Why Ron Rivest’s work underpins the internet and blockchains 08:10 Before public-key cryptography, there was no theory of security 11:05 The open problem that led to RSA 13:55 The night Ron Rivest discovered the core idea behind RSA 17:55 Why digital signatures were the real breakthrough 19:14 The RSA challenge — and a prediction that was off by quadrillions of years 28:33 How government pressure shaped cryptographic standards 30:36 Designing the hash functions that made digital signatures practical 33:26 The Fiat–Shamir transformation, explained 38:04 Building a cryptography company before the web existed 42:31 Will quantum computers ever become powerful enough to break RSA? 48:02 The cryptography securing the internet, elections, and everyday life 50:11 What surprised Dan: Quantum giveth and quantum taketh away

a16z crypto

62,339 Aufrufe • vor 23 Tagen

First Principles Ep. 4 with Alvin Roth Long before crypto made coordination programmable, Nobel Prize winner Alvin Roth was designing markets where coordination could save lives, matching doctors to hospitals, students to schools, and kidney donors to the patients who need them. Roth explains why markets are not just natural forces, but engineered systems; why the details of timing, congestion, incentives, and trust can make or break a marketplace; and why some of the most important markets are the ones where simply exchanging money can’t do the work. Hosted by Tim Roughgarden with Scott Kominers. 00:00 Intro: Why market design matters 04:18 The economist as engineer 08:09 When theory meets the real world 07:02 Fixing the medical residency match 15:32 Why markets unravel 18:22 Redesigning NYC high school admissions 28:05 The hidden problem of congestion 34:47 How kidney exchange saves lives 45:26 How the internet changed market design 48:25 Airbnb, Uber and smarter marketplaces 51:28 Repugnant transactions and moral economics 53:32 When markets need social support 54:32 The unexpected effects of criminalizing surrogacy 01:04:58 Preference signals and the job market 01:18:53 A broken market: resettling refugees and other migrants
1:25:25

First Principles Ep. 4 with Alvin Roth Long before crypto made coordination programmable, Nobel Prize winner Alvin Roth was designing markets where coordination could save lives, matching doctors to hospitals, students to schools, and kidney donors to the patients who need them. Roth explains why markets are not just natural forces, but engineered systems; why the details of timing, congestion, incentives, and trust can make or break a marketplace; and why some of the most important markets are the ones where simply exchanging money can’t do the work. Hosted by Tim Roughgarden with Scott Kominers. 00:00 Intro: Why market design matters 04:18 The economist as engineer 08:09 When theory meets the real world 07:02 Fixing the medical residency match 15:32 Why markets unravel 18:22 Redesigning NYC high school admissions 28:05 The hidden problem of congestion 34:47 How kidney exchange saves lives 45:26 How the internet changed market design 48:25 Airbnb, Uber and smarter marketplaces 51:28 Repugnant transactions and moral economics 53:32 When markets need social support 54:32 The unexpected effects of criminalizing surrogacy 01:04:58 Preference signals and the job market 01:18:53 A broken market: resettling refugees and other migrants

a16z crypto

125,607 Aufrufe • vor 1 Monat

Context Rot! What is it? What does it mean? New video breaking down Chroma's latest paper and applying it to interacting with AI 00:00 - What is Context? 04:44 - The Flaw of "Needle in a Haystack" 10:22 - How Distractors Degrade AI Performance 14:31 - The Impact of Prompt Structure 25:55 - Why Unrelated Needles Are Easier to Find 31:02 - The Unreliability of Text Replication 37:03 - Actionable Takeaways for Better Prompting
40:35

Context Rot! What is it? What does it mean? New video breaking down Chroma's latest paper and applying it to interacting with AI 00:00 - What is Context? 04:44 - The Flaw of "Needle in a Haystack" 10:22 - How Distractors Degrade AI Performance 14:31 - The Impact of Prompt Structure 25:55 - Why Unrelated Needles Are Easier to Find 31:02 - The Unreliability of Text Replication 37:03 - Actionable Takeaways for Better Prompting

matt palmer

22,103 Aufrufe • vor 11 Monaten

Hey Siri, why are American kids so sick? Aaron Siri on DarkHorse with Bret Weinstein : 00:00:00 Welcome Aaron Siri 00:05:22 Vaccines: Then vs. Now 00:18:04 What We Don’t Know About Childhood Vaccines 00:33:45 Financial Incentives in Vaccine Development 00:52:02 Polio Vaccine: Unintended Consequences 00:55:39 The Polio Story as a Red Pill 01:25:32 HIV Origins & Parallels to COVID 01:45:04 Rethinking the Spanish Flu 01:47:31 When Interventions Cause Harm 01:49:40 Measles Mortality Data 01:56:07 Health Before Vaccination 02:03:17 Defining “Vaccine” 02:05:01 Why Hep B Remains on the Schedule 02:11:29 Autism & Vaccines 02:20:02 The Tyranny of Vaccine Mandates 02:35:27 Scientific vs. Legal Fraud 02:53:00 Trump Administration & the Childhood Schedule 03:04:38 Think of Vaccines as a Product
3:08:34

Hey Siri, why are American kids so sick? Aaron Siri on DarkHorse with Bret Weinstein : 00:00:00 Welcome Aaron Siri 00:05:22 Vaccines: Then vs. Now 00:18:04 What We Don’t Know About Childhood Vaccines 00:33:45 Financial Incentives in Vaccine Development 00:52:02 Polio Vaccine: Unintended Consequences 00:55:39 The Polio Story as a Red Pill 01:25:32 HIV Origins & Parallels to COVID 01:45:04 Rethinking the Spanish Flu 01:47:31 When Interventions Cause Harm 01:49:40 Measles Mortality Data 01:56:07 Health Before Vaccination 02:03:17 Defining “Vaccine” 02:05:01 Why Hep B Remains on the Schedule 02:11:29 Autism & Vaccines 02:20:02 The Tyranny of Vaccine Mandates 02:35:27 Scientific vs. Legal Fraud 02:53:00 Trump Administration & the Childhood Schedule 03:04:38 Think of Vaccines as a Product

The DarkHorse Podcast

69,778 Aufrufe • vor 7 Monaten

Full Shard Podcast: Episode 12 - SUPERNOVA When technology becomes fast enough, it disappears. In this episode, Beniamin Mincu |🇺🇸/acc and Adrian Dobrita step back from specs and roadmaps to talk about what Supernova really represents. From redefining what a blockchain is, to why speed changes human behavior, to how AI agents may soon need blockchains to coordinate, this is a big-picture conversation about building systems that can keep up with the future. Not just about how it works. But about why it matters. 00:00 - The Importance of Speed 8:44 - What is Actually a Blockchain? 11:54 - How to Build the Future 18:16 - How Supernova Works 21:55 - The New Consensus and Execution Mechanisms 32:46 - What Supernova Enables 41:41 - ZK and Privacy on MultiversX 43:14 - Building a Blockchain From Scratch 47:48 - Supernova Launch Date 49:04 - How to Deal with Challenges 51:15 - Ending Thoughts
52:34

Full Shard Podcast: Episode 12 - SUPERNOVA When technology becomes fast enough, it disappears. In this episode, Beniamin Mincu |🇺🇸/acc and Adrian Dobrita step back from specs and roadmaps to talk about what Supernova really represents. From redefining what a blockchain is, to why speed changes human behavior, to how AI agents may soon need blockchains to coordinate, this is a big-picture conversation about building systems that can keep up with the future. Not just about how it works. But about why it matters. 00:00 - The Importance of Speed 8:44 - What is Actually a Blockchain? 11:54 - How to Build the Future 18:16 - How Supernova Works 21:55 - The New Consensus and Execution Mechanisms 32:46 - What Supernova Enables 41:41 - ZK and Privacy on MultiversX 43:14 - Building a Blockchain From Scratch 47:48 - Supernova Launch Date 49:04 - How to Deal with Challenges 51:15 - Ending Thoughts

Full Shard Podcast

27,714 Aufrufe • vor 6 Monaten

First Principles Ep. 6 with Shafi Goldwasser What if you could prove something is true without revealing why it’s true? Turing Award winner Shafi Goldwasser tells the origin story of zero-knowledge proofs: how a playful question about playing poker securely over the telephone led her, Silvio Micali, and Charles Rackoff to rethink what a mathematical proof could be. Goldwasser traces those ideas through interactive proofs, the sum-check protocol, SNARKs, and the systems now used to verify computation and preserve privacy on blockchains. She also reflects on why breakthrough ideas are often rejected at first, how toy problems can produce foundational theories, and whether AI systems should have to prove their answers. Hosted by Tim Roughgarden with Justin Thaler 00:00 Intro 03:36 How mental poker inspired zero-knowledge proofs: Proving that something is true without revealing the underlying information 06:30 The simulation paradigm and the meaning of “zero knowledge” 08:33 Why the original paper was repeatedly rejected 10:33 How interactive proofs became more powerful than conventional proofs 13:36 The road to modern SNARKs 19:02 Why the sum-check protocol is so useful for verifiable computation 25:31 Why many so-called “zk proofs” are not actually zero knowledge 34:06 Why toy examples, playfulness, and narratives can produce deep theory 37:23 The role of rigor and computational assumptions in cryptography 42:44 Applying zero-knowledge proofs to law, evidence, and secret software 45:23 Training AI systems to provide proofs alongside their answers 54:27 Why genuinely new ideas are often difficult for experts to recognize
57:01

First Principles Ep. 6 with Shafi Goldwasser What if you could prove something is true without revealing why it’s true? Turing Award winner Shafi Goldwasser tells the origin story of zero-knowledge proofs: how a playful question about playing poker securely over the telephone led her, Silvio Micali, and Charles Rackoff to rethink what a mathematical proof could be. Goldwasser traces those ideas through interactive proofs, the sum-check protocol, SNARKs, and the systems now used to verify computation and preserve privacy on blockchains. She also reflects on why breakthrough ideas are often rejected at first, how toy problems can produce foundational theories, and whether AI systems should have to prove their answers. Hosted by Tim Roughgarden with Justin Thaler 00:00 Intro 03:36 How mental poker inspired zero-knowledge proofs: Proving that something is true without revealing the underlying information 06:30 The simulation paradigm and the meaning of “zero knowledge” 08:33 Why the original paper was repeatedly rejected 10:33 How interactive proofs became more powerful than conventional proofs 13:36 The road to modern SNARKs 19:02 Why the sum-check protocol is so useful for verifiable computation 25:31 Why many so-called “zk proofs” are not actually zero knowledge 34:06 Why toy examples, playfulness, and narratives can produce deep theory 37:23 The role of rigor and computational assumptions in cryptography 42:44 Applying zero-knowledge proofs to law, evidence, and secret software 45:23 Training AI systems to provide proofs alongside their answers 54:27 Why genuinely new ideas are often difficult for experts to recognize

a16z crypto

144,449 Aufrufe • vor 8 Tagen

My full conversation with Yat Siu, Chairman and Co-Founder of Animoca Brands "The people who win are the ones who learn to be divergent." 00:00:00 - Intro 00:01:43 - Growing Up in Austria: Music, Coding & Being an Outsider 00:07:40 - The Iron Curtain, Capitalism & What Shaped Yat's Worldview 00:14:53 - Building a Business from the Wreckage 00:18:43 - 3D Graphics, Early VR & the Metaverse Before Its Time 00:24:00 - Dot-Com Boom, Outblaze & Selling to IBM 00:28:17 - MUDs, Virtual Swords & the Origins of Digital Ownership 00:33:59 - The Animoca Origin Story 00:38:29 - Getting Deplatformed by Apple & Pivoting to Android 00:40:54 - CryptoKitties, NFTs & Going All-In During Crypto Winter 00:47:50 - Investing in De-Extinction: Colossal Biosciences & Real-Life Jurassic Park 00:50:41 - Agents, Tools & the 1% Problem 00:56:37 - Why Tokens & Blockchain Matter for AI 01:04:30 - Advice for the Young: Embrace Uncertainty & Be Divergent
1:10:04

My full conversation with Yat Siu, Chairman and Co-Founder of Animoca Brands "The people who win are the ones who learn to be divergent." 00:00:00 - Intro 00:01:43 - Growing Up in Austria: Music, Coding & Being an Outsider 00:07:40 - The Iron Curtain, Capitalism & What Shaped Yat's Worldview 00:14:53 - Building a Business from the Wreckage 00:18:43 - 3D Graphics, Early VR & the Metaverse Before Its Time 00:24:00 - Dot-Com Boom, Outblaze & Selling to IBM 00:28:17 - MUDs, Virtual Swords & the Origins of Digital Ownership 00:33:59 - The Animoca Origin Story 00:38:29 - Getting Deplatformed by Apple & Pivoting to Android 00:40:54 - CryptoKitties, NFTs & Going All-In During Crypto Winter 00:47:50 - Investing in De-Extinction: Colossal Biosciences & Real-Life Jurassic Park 00:50:41 - Agents, Tools & the 1% Problem 00:56:37 - Why Tokens & Blockchain Matter for AI 01:04:30 - Advice for the Young: Embrace Uncertainty & Be Divergent

JackGK

28,960 Aufrufe • vor 5 Monaten

François Chollet (François Chollet) has spent years asking a different question than most of the AI world. Instead of scaling what already works, he’s trying to understand what intelligence actually is and how to build it from first principles. In this episode of the Lightcone Podcast, he traces that path from his early work on deep learning to the creation of the ARC Prize, and the launch of ARC V3, a new benchmark designed to measure something deeper than performance: the ability to learn, adapt, and reason efficiently in entirely new environments. He explains why today’s systems may be hitting limits, what recent breakthroughs really mean, and why reaching true general intelligence may require a fundamentally different approach. 00:00 - AGI by 2030? 00:31 - Introducing Ndea: A New Path Beyond Deep Learning 01:08 - A New ML Paradigm 01:30 - Replacing neural nets with compact symbolic programs 03:04 - Why Ndea Isn’t Competing With Coding Agents 05:20 - Why Everyone Might Be Wrong About Scaling LLMs 07:22 - Why Coding Agents Suddenly Work So Well 08:50 - The Limits of LLMs in Non-Verifiable Domains 10:48 - What AGI Actually Means (And Why Most Definitions Are Wrong) 13:30 - Why Deep Learning Hits a Wall 14:00 - ARC’s Origin Story 18:20 - ARC Benchmarks Explained: From V1 to V3 22:49 - The RL Loop Powering Coding Agents Today 27:03 - ARC-AGI V3: Measuring “Agentic Intelligence” 31:14 - Inside the ARC Game Studio 35:31 - Could AGI Fit in 10,000 Lines of Code? 44:01 - Building Ndea: From Idea to Compounding Research Stack 46:46 - The Future of ARC: Benchmarks That Evolve With AI 47:21 - Why There’s Still Huge Opportunity for New AI Paradigms 53:37 - How to Build a Breakout Open Source Project - Lessons From Keras 56:39 - Advice For How To Think About AI
57:23

François Chollet (François Chollet) has spent years asking a different question than most of the AI world. Instead of scaling what already works, he’s trying to understand what intelligence actually is and how to build it from first principles. In this episode of the Lightcone Podcast, he traces that path from his early work on deep learning to the creation of the ARC Prize, and the launch of ARC V3, a new benchmark designed to measure something deeper than performance: the ability to learn, adapt, and reason efficiently in entirely new environments. He explains why today’s systems may be hitting limits, what recent breakthroughs really mean, and why reaching true general intelligence may require a fundamentally different approach. 00:00 - AGI by 2030? 00:31 - Introducing Ndea: A New Path Beyond Deep Learning 01:08 - A New ML Paradigm 01:30 - Replacing neural nets with compact symbolic programs 03:04 - Why Ndea Isn’t Competing With Coding Agents 05:20 - Why Everyone Might Be Wrong About Scaling LLMs 07:22 - Why Coding Agents Suddenly Work So Well 08:50 - The Limits of LLMs in Non-Verifiable Domains 10:48 - What AGI Actually Means (And Why Most Definitions Are Wrong) 13:30 - Why Deep Learning Hits a Wall 14:00 - ARC’s Origin Story 18:20 - ARC Benchmarks Explained: From V1 to V3 22:49 - The RL Loop Powering Coding Agents Today 27:03 - ARC-AGI V3: Measuring “Agentic Intelligence” 31:14 - Inside the ARC Game Studio 35:31 - Could AGI Fit in 10,000 Lines of Code? 44:01 - Building Ndea: From Idea to Compounding Research Stack 46:46 - The Future of ARC: Benchmarks That Evolve With AI 47:21 - Why There’s Still Huge Opportunity for New AI Paradigms 53:37 - How to Build a Breakout Open Source Project - Lessons From Keras 56:39 - Advice For How To Think About AI

Y Combinator

151,649 Aufrufe • vor 4 Monaten

🎙️ New The Edge Podcast 🪷 The Next Battle In DeFi Lending Is What's Inside Your Vault | DeFi Frontier 0:00 - Intro 2:22 - The team behind Lotus 3:45 - The problem with shared pool design 7:23 - The “missing middle” in DeFi lending 9:31 - The problem with isolated markets 11:17 - What does Lotus aim to solve? 13:40 - The elevator pitch for Lotus 19:54 - How tranching plays a role in Lotus 24:00 - Demo of Lotus and how it works 28:56 - Credora ratings in Lotus 33:19 - Earning more with productive debt 37:52 - Productive debt dampens rate volatility 42:13 - How Lotus differs handling bad debt 45:08 - Who will use Lotus? 48:04 - What uses could Lotus unlock? 50:49 - Security, audits, and lessons from rsETH 57:02 - Closing + Lotus launch 🙏 Thanks to Lotus Protocol Founder David Reising for joining us!
59:24

🎙️ New The Edge Podcast 🪷 The Next Battle In DeFi Lending Is What's Inside Your Vault | DeFi Frontier 0:00 - Intro 2:22 - The team behind Lotus 3:45 - The problem with shared pool design 7:23 - The “missing middle” in DeFi lending 9:31 - The problem with isolated markets 11:17 - What does Lotus aim to solve? 13:40 - The elevator pitch for Lotus 19:54 - How tranching plays a role in Lotus 24:00 - Demo of Lotus and how it works 28:56 - Credora ratings in Lotus 33:19 - Earning more with productive debt 37:52 - Productive debt dampens rate volatility 42:13 - How Lotus differs handling bad debt 45:08 - Who will use Lotus? 48:04 - What uses could Lotus unlock? 50:49 - Security, audits, and lessons from rsETH 57:02 - Closing + Lotus launch 🙏 Thanks to Lotus Protocol Founder David Reising for joining us!

DeFi Dad ⟠ defidad.eth

21,736 Aufrufe • vor 3 Monaten

Full interview with the founder and CEO of Propr Lou sits down with me to discuss Hyperliquid and the future of prop trading Timestamps: 00:00:00 Intro 00:00:16 Meet Louis: from TradFi to crypto 00:01:52 How the crypto market has changed 00:03:49 Why build on Hyperliquid 00:06:52 How propr got started 00:08:40 Scaling a lean team (and the grind) 00:12:24 What a prop firm actually is 00:15:33 Agents, bots, and the future of trading 00:18:32 What Louis is most proud of 00:21:02 Thinking about the token and TGE 00:24:14 Where propr goes in 5 years 00:25:57 Best and worst trades 00:27:50 Life outside crypto + shoutouts
29:41

Full interview with the founder and CEO of Propr Lou sits down with me to discuss Hyperliquid and the future of prop trading Timestamps: 00:00:00 Intro 00:00:16 Meet Louis: from TradFi to crypto 00:01:52 How the crypto market has changed 00:03:49 Why build on Hyperliquid 00:06:52 How propr got started 00:08:40 Scaling a lean team (and the grind) 00:12:24 What a prop firm actually is 00:15:33 Agents, bots, and the future of trading 00:18:32 What Louis is most proud of 00:21:02 Thinking about the token and TGE 00:24:14 Where propr goes in 5 years 00:25:57 Best and worst trades 00:27:50 Life outside crypto + shoutouts

ryandcrypto

42,836 Aufrufe • vor 2 Monaten

How does math research change when the cost of trying your first dumb idea goes to zero? Daniel Litt joins Greg Burnham and Anson Ho to discuss what today’s models can and can’t do in math, and how far they are from doing high-quality research. 0:00:00 What's the hardest math problem AI can solve today? 00:16:08 How helpful are today’s AI models for math research? 00:23:36 Junk papers, LLM-generated proofs, and the refereeing crisis 00:27:21 AI enables searching through problems at scale 00:33:49 When will AI be good enough to publish in top math journals? 00:42:15 What are the returns to intelligence? 00:59:50 Will AI solve Millennium problems? 01:11:54 Is math full of low-hanging fruit? 01:18:47 How Daniel has adapted his professional life to AI progress 01:25:28 What do AI math benchmarks actually measure? 01:33:05 Designing the Open Problems benchmark 01:56:35 Do mathematicians believe heuristic arguments about conjectures? 02:01:24 What if FrontierMath: Open Problems gets solved? 02:06:53 Is AI on the cusp of accelerating math progress?
2:10:59

How does math research change when the cost of trying your first dumb idea goes to zero? Daniel Litt joins Greg Burnham and Anson Ho to discuss what today’s models can and can’t do in math, and how far they are from doing high-quality research. 0:00:00 What's the hardest math problem AI can solve today? 00:16:08 How helpful are today’s AI models for math research? 00:23:36 Junk papers, LLM-generated proofs, and the refereeing crisis 00:27:21 AI enables searching through problems at scale 00:33:49 When will AI be good enough to publish in top math journals? 00:42:15 What are the returns to intelligence? 00:59:50 Will AI solve Millennium problems? 01:11:54 Is math full of low-hanging fruit? 01:18:47 How Daniel has adapted his professional life to AI progress 01:25:28 What do AI math benchmarks actually measure? 01:33:05 Designing the Open Problems benchmark 01:56:35 Do mathematicians believe heuristic arguments about conjectures? 02:01:24 What if FrontierMath: Open Problems gets solved? 02:06:53 Is AI on the cusp of accelerating math progress?

Epoch AI

178,703 Aufrufe • vor 6 Monaten

My conversation with OpenAI co-founder Greg Brockman This is the most detailed first-person account of the 72 hours after Sam Altman was fired. We also go deep on what comes next: the global race to AGI, why ChatGPT stopped showing reasoning, how much of OpenAI's own code is now written by AI ("it's hard to know what percent is not"), and the untold story of how OpenAI actually started in 2015. 00:00:00 Introduction 00:00:49 Meeting Sam Altman and Starting OpenAI 00:02:40 Building the Founding Team 00:04:25 DeepMind's Lead Over OpenAI 00:04:54 Changing OpenAI to a For-Profit Model 00:06:05 Breakthrough Moments at OpenAI 00:08:22 What Dota 2 Meant for OpenAI 00:10:04 Reasoning Versus Prediction 00:11:59 Tensions Grow at OpenAI 00:15:44 Sam Altman's Firing 00:17:49 Greg Quits OpenAI 00:19:56 Sam Explores Deal with Microsoft's Satya 00:20:28 Petition for Altman's Return 00:23:43 Ilya Sutskever Leaves OpenAI 00:24:59 Lessons Learned after Sam Ousting 00:28:22 The Thing Ilya Said that Greg Can't Forget 00:32:22 Is AI Going Parabolic? 00:33:24 How Much of OpenAI's Code is Written by AI? 00:36:21 Do AI Chatbots Tell Us What We Want to Hear? 00:38:06 The Global AI Race to Reach AGI 00:38:40 What Happens if US Doesn't Reach AGI First? 00:39:49 Are Countries Stealing AI Advancements? 00:40:38 Why ChatGPT No Longer Shows Reasoning 00:41:47 The Finite Constraints of Compute 00:43:38 On Investing Early in Data Centers 00:46:31 The Future of Data Center Specialization 00:47:52 How to Decide Whose Queries to Serve 00:49:08 OpenAI on Consumer vs Enterprise Models 00:53:05 Data Centers in Space? 01:00:56 What Should AI Regulation Look Like? 01:04:33 The Future of AI-Powered Entrepreneurship 01:04:44 AI and Job Loss 01:07:15 The Skills Young People Should Invest In 01:11:30 What Does Success Look Like For You? Full episode on X below. Also find it on: • YouTube: • Spotify: • Apple:
1:11:41

My conversation with OpenAI co-founder Greg Brockman This is the most detailed first-person account of the 72 hours after Sam Altman was fired. We also go deep on what comes next: the global race to AGI, why ChatGPT stopped showing reasoning, how much of OpenAI's own code is now written by AI ("it's hard to know what percent is not"), and the untold story of how OpenAI actually started in 2015. 00:00:00 Introduction 00:00:49 Meeting Sam Altman and Starting OpenAI 00:02:40 Building the Founding Team 00:04:25 DeepMind's Lead Over OpenAI 00:04:54 Changing OpenAI to a For-Profit Model 00:06:05 Breakthrough Moments at OpenAI 00:08:22 What Dota 2 Meant for OpenAI 00:10:04 Reasoning Versus Prediction 00:11:59 Tensions Grow at OpenAI 00:15:44 Sam Altman's Firing 00:17:49 Greg Quits OpenAI 00:19:56 Sam Explores Deal with Microsoft's Satya 00:20:28 Petition for Altman's Return 00:23:43 Ilya Sutskever Leaves OpenAI 00:24:59 Lessons Learned after Sam Ousting 00:28:22 The Thing Ilya Said that Greg Can't Forget 00:32:22 Is AI Going Parabolic? 00:33:24 How Much of OpenAI's Code is Written by AI? 00:36:21 Do AI Chatbots Tell Us What We Want to Hear? 00:38:06 The Global AI Race to Reach AGI 00:38:40 What Happens if US Doesn't Reach AGI First? 00:39:49 Are Countries Stealing AI Advancements? 00:40:38 Why ChatGPT No Longer Shows Reasoning 00:41:47 The Finite Constraints of Compute 00:43:38 On Investing Early in Data Centers 00:46:31 The Future of Data Center Specialization 00:47:52 How to Decide Whose Queries to Serve 00:49:08 OpenAI on Consumer vs Enterprise Models 00:53:05 Data Centers in Space? 01:00:56 What Should AI Regulation Look Like? 01:04:33 The Future of AI-Powered Entrepreneurship 01:04:44 AI and Job Loss 01:07:15 The Skills Young People Should Invest In 01:11:30 What Does Success Look Like For You? Full episode on X below. Also find it on: • YouTube: • Spotify: • Apple:

Shane Parrish

450,952 Aufrufe • vor 3 Monaten

Diffusion is the foundational ML framework behind state-of-the-art AI image and video generation, including Sora, Midjourney and Google Veo. In this episode of Decoded, Ankit Gupta sits down with Francois Chaubard to discuss how diffusion works, walk through a code sample, and explain why everyone training models should understand it. 00:00 Intro 00:33 What is diffusion? 02:50 What are applications of diffusion today? 04:06 Key innovations 07:01 Code examples 19:25 The "squint test" 22:27 Other areas diffusion is widely accessible 24:49 Outro
27:10

Diffusion is the foundational ML framework behind state-of-the-art AI image and video generation, including Sora, Midjourney and Google Veo. In this episode of Decoded, Ankit Gupta sits down with Francois Chaubard to discuss how diffusion works, walk through a code sample, and explain why everyone training models should understand it. 00:00 Intro 00:33 What is diffusion? 02:50 What are applications of diffusion today? 04:06 Key innovations 07:01 Code examples 19:25 The "squint test" 22:27 Other areas diffusion is widely accessible 24:49 Outro

Y Combinator

74,604 Aufrufe • vor 6 Monaten