{"success":true,"count":8,"items":[{"videoId":"BD3vLtWhT5A","chunkIndex":0,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 1 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["artificial-intelligence","machine-learning","technology-trends","software-industry","automation","future-of-work","jobs","enterprise-software"],"normalizedKeywords":["기술 트렌드","비즈니스·전략","엔지니어링"],"targetAudience":[{"who":"창업자","why":"AI가 소프트웨어와 B2B 시장에 미칠 변화를 사업 기회 관점에서 볼 수 있음"},{"who":"엔지니어","why":"AI 개발 도구와 소프트웨어 개발 방식의 변화를 이해하는 데 유용함"},{"who":"지식노동자","why":"자기 일의 어떤 부분이 AI에 의해 바뀔지 현실적으로 점검할 수 있음"},{"who":"투자자","why":"AI 가치사슬과 시장 성숙도를 읽는 관점이 담겨 있음"}],"normalizedAudience":["창업자·스타트업","엔지니어·개발자","투자자·VC"],"summary":"이 대화는 AI를 과장도 과소평가도 하지 말고, 인터넷·모바일급의 거대한 변화로 보되 아직은 1997년쯤의 초기 단계로 이해해야 한다는 관점을 중심으로 전개된다. Benedict Evans는 많은 사람이 AI의 영향 범위를 너무 빨리 일반화하고 있다고 지적하며, 실제로는 도구의 성숙도와 채택 수준이 매우 들쭉날쭉하다고 말한다.\n\n또한 그는 '당장 일자리가 대거 사라진다'는 식의 단순한 예측을 비판하고, 기술이 기존 일을 자동화하는 동시에 새로운 일을 만들어낸다는 반복 패턴을 강조한다. 소프트웨어 분야는 이미 변화를 체감하는 단계지만, 법률·회계 같은 다른 전문직은 아직 어디에 어떻게 적용될지 탐색 중이며, 따라서 중요한 것은 거부가 아니라 직접 써보고 이해하는 태도라고 주장한다.","insights":["AI는 혁명급이지만 아직은 초기 채택 곡선의 한복판이다.","일자리는 한꺼번에 사라지지 않고, 자동화와 신직무가 같이 온다.","도구의 영향은 직업별로 다르게 드러나며, 예측은 항상 거칠다.","AI를 싫어하는 태도보다 직접 써보는 태도가 미래 적응에 유리하다.","가치는 모델 자체보다 이를 현업에 연결하는 적용층에서 생긴다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c0:24-37","startSegmentIndex":24,"endSegmentIndex":37,"startTime":166.48,"endTime":315.759,"durationSeconds":149.3,"preview":"AI는 1997년 단계","mustSee":true},{"clipId":"BD3vLtWhT5A:c0:50-63","startSegmentIndex":50,"endSegmentIndex":63,"startTime":406.72,"endTime":516.24,"durationSeconds":109.5,"preview":"소프트웨어부터 변한다","mustSee":false}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T11:58:04.527Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":1,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 2 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["ai","automation","consulting","enterprise-software","productivity","labor-market","digital-transformation","business-strategy","platform-shifts"],"normalizedKeywords":["비즈니스·전략","커리어·성장","기술 트렌드"],"targetAudience":[{"who":"창업자","why":"AI가 실제로 어떤 업무를 대체·증폭하는지 사업 관점에서 보게 해줌"},{"who":"투자자","why":"AI 도입이 인력 감축보다 조직 재설계와 수요 확장으로 이어질 수 있음을 보여줌"},{"who":"지식노동자","why":"자동화 시대에도 일이 어떻게 재구성되는지 이해하는 데 유용함"}],"normalizedAudience":["창업자·스타트업","투자자·VC","지식노동자 일반"],"summary":"이 영상은 AI가 단순히 사람을 대체해 일자리를 줄인다는 통념을 비판하면서, 실제로는 업무의 '작업(task)'과 '일(job)'이 다르기 때문에 자동화의 결과가 훨씬 복잡하다고 주장한다. 컨설팅, PE, 소프트웨어, 회계 같은 사례를 들어, 기술이 어떤 부분의 비용을 낮추면 같은 일을 더 적게 하게 되는 것이 아니라 오히려 더 많은 일과 새로운 수요가 생기는 경우가 많다고 설명한다.\n\n핵심 메시지는 '자동화=인력 감소'가 아니라 '자동화=재배치와 확장'이라는 점이다. AI 기업들조차 인력을 늘리고 있으며, 가장 가치 있는 일은 코드를 쓰는 행위보다 무엇을 만들지, 누구에게 팔지, 조직 내부를 어떻게 바꿀지 결정하는 고차원 문제라는 점을 강조한다. 결국 AI 시대에도 남는 것은 단순 생산이 아니라 문제 정의, 조직 설계, 고객 이해 같은 상위 수준의 일이다.","insights":["작업이 쉬워져도 일 전체가 사라지지는 않는다.","자동화는 인력 감축보다 수요 확장을 먼저 만든다.","가장 비싼 일은 '무엇을 할지'를 정하는 일이다.","기술은 코드보다 조직과 업무 흐름을 바꿀 때 진짜 효과가 난다.","AI 기업도 사람을 줄이기보다 오히려 더 뽑고 있다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c1:2-15","startSegmentIndex":2,"endSegmentIndex":15,"startTime":628.16,"endTime":747.12,"durationSeconds":119,"preview":"컨설팅이 사라지지 않는 이유","mustSee":false},{"clipId":"BD3vLtWhT5A:c1:16-28","startSegmentIndex":16,"endSegmentIndex":28,"startTime":747.12,"endTime":840.32,"durationSeconds":93.2,"preview":"태스크와 잡의 차이","mustSee":true},{"clipId":"BD3vLtWhT5A:c1:29-41","startSegmentIndex":29,"endSegmentIndex":41,"startTime":840.32,"endTime":928.72,"durationSeconds":88.4,"preview":"기술이 늘린 일","mustSee":false},{"clipId":"BD3vLtWhT5A:c1:42-57","startSegmentIndex":42,"endSegmentIndex":57,"startTime":928.72,"endTime":1033.919,"durationSeconds":105.2,"preview":"아웃소싱된 판단","mustSee":false},{"clipId":"BD3vLtWhT5A:c1:58-72","startSegmentIndex":58,"endSegmentIndex":72,"startTime":1033.919,"endTime":1144.08,"durationSeconds":110.2,"preview":"AI 시대의 일자리","mustSee":false}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:56:00.449Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":2,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 3 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["artificial-intelligence","ai-adoption","enterprise-software","digital-transformation","technology-trends","automation","productivity","business-change"],"normalizedKeywords":["비즈니스·전략","기술 트렌드"],"targetAudience":[{"who":"창업자","why":"AI 도입이 실제로 얼마나 빨리 확산되고 사업이 바뀌는지 감 잡는 데 유용함"},{"who":"프로덕트 매니저","why":"신기술이 조직과 제품에 스며드는 속도와 현실적 제약을 이해할 수 있음"},{"who":"지식노동자","why":"AI가 일자리를 즉시 대체하기보다 업무 구조를 바꾸는 방식에 대한 통찰을 줌"}],"normalizedAudience":["창업자·스타트업","프로덕트 매니저·기획자","지식노동자 일반"],"summary":"이 영상은 AI가 기존 기술 혁신과 비교해 정말 '완전히 다른가'를 차분하게 따진다. 화자는 인터넷, PC, 전기, 바코드 같은 과거 기술도 이미 엄청난 변화를 만들었지만, 실제 확산과 업무 재편은 생각보다 오래 걸렸다고 강조한다. AI 역시 도입 속도는 빠를 수 있지만, 대기업의 구매·전환 사이클과 산업별 관성 때문에 업무와 조직은 3~10년에 걸쳐 서서히 바뀐다고 본다.\n\n동시에 그는 AGI나 초지능 같은 용어가 계속 재정의되고 있으며, 현재로서는 인간 수준 지능의 도달 여부를 알 수 없다고 말한다. 다만 그런 논쟁과 무관하게, 지금의 AI는 이미 매우 유용한 변혁적 기술이며 앞으로 10년 동안 세계를 크게 바꿀 것이라는 점은 분명하다고 정리한다.","insights":["혁신의 속도는 기술보다 기존 인프라에 좌우된다.","AI는 즉시 대체보다 점진적 재설계에 가깝다.","기업은 새 기술을 알아도 조직 전환은 느리다.","AGI 논쟁보다 중요한 건 지금의 실사용 가치다.","새 기술의 이름은 늘 '지금 막 되는 것'으로 바뀐다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c2:1-12","startSegmentIndex":1,"endSegmentIndex":12,"startTime":1201.51,"endTime":1340.72,"durationSeconds":139.2,"preview":"AI 확산이 빠른 이유","mustSee":false},{"clipId":"BD3vLtWhT5A:c2:13-20","startSegmentIndex":13,"endSegmentIndex":20,"startTime":1340.72,"endTime":1398.48,"durationSeconds":57.8,"preview":"변화는 늘 늦게 온다","mustSee":false},{"clipId":"BD3vLtWhT5A:c2:21-28","startSegmentIndex":21,"endSegmentIndex":28,"startTime":1398.48,"endTime":1464.64,"durationSeconds":66.2,"preview":"과거 혁신의 반복","mustSee":false},{"clipId":"BD3vLtWhT5A:c2:29-42","startSegmentIndex":29,"endSegmentIndex":42,"startTime":1464.64,"endTime":1562.799,"durationSeconds":98.2,"preview":"인터넷이 바꾼 정보","mustSee":false},{"clipId":"BD3vLtWhT5A:c2:43-54","startSegmentIndex":43,"endSegmentIndex":54,"startTime":1562.799,"endTime":1744.96,"durationSeconds":182.2,"preview":"AGI보다 중요한 것","mustSee":true}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:56:38.707Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":3,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 4 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["ai","foundation-models","llm","software-platforms","cloud-computing","market-structure","monetization","big-tech","technology-strategy"],"normalizedKeywords":["비즈니스·전략","기술 트렌드"],"targetAudience":[{"who":"창업자","why":"AI 모델과 애플리케이션 층의 가치 배분을 이해하는 데 직접적이다."},{"who":"투자자","why":"기초모델과 앱 레이어 중 어디에 경제적 해자가 생길지 판단하는 데 유용하다."},{"who":"프로덕트 매니저","why":"챗봇 UX와 앱 중심 UX의 차이가 제품 전략을 어떻게 바꾸는지 볼 수 있다."}],"normalizedAudience":["창업자·스타트업","투자자·VC","프로덕트 매니저·기획자"],"summary":"이 구간은 AI가 단순히 기존 시장을 대체하는 수준이 아니라, 기업의 총주소시장(TAM)과 가능한 회사 규모 자체를 키우는 플랫폼 전환이라는 관점에서 출발한다. 화자는 과거 메인프레임→PC→인터넷→모바일로 이어진 확장처럼, AI도 더 많은 업무와 경제 영역을 자동화하며 기회를 넓힐 것이라고 본다.\n\n동시에 그는 기초모델 회사의 수익성이 과대평가될 수 있다고 의심한다. 챗봇이 최종 UX가 아니고, 실제 가치는 수많은 앱·워크플로우 위에 쌓인다면 모델은 AWS처럼 비교적 표준화된 인프라가 되어 가격 경쟁에 노출될 수 있다는 것이다. 반대로 윈도우처럼 강한 락인과 네트워크 효과가 생기지 않는다면, 가치의 대부분은 모델보다 상위 레이어로 이동할 가능성이 크다고 주장한다.","insights":["AI는 기존 시장을 잠식하기보다 TAM 자체를 키운다.","모델이 곧 제품이 아니면 가치의 중심은 앱 레이어로 이동한다.","네트워크 효과가 약한 AI 모델은 장기적으로 가격 압박을 받기 쉽다.","플랫폼 전환기에는 현재의 승자보다 상위 레이어의 새 승자를 봐야 한다.","초기 과열 가격은 정상화되고, 그 뒤의 균형이 진짜 수익성을 결정한다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c3:1-5","startSegmentIndex":1,"endSegmentIndex":5,"startTime":1801.11,"endTime":1823.36,"durationSeconds":22.3,"preview":"AI가 키우는 시장","mustSee":false},{"clipId":"BD3vLtWhT5A:c3:7-11","startSegmentIndex":7,"endSegmentIndex":11,"startTime":1892.72,"endTime":1991.2,"durationSeconds":98.5,"preview":"일자리와 가치의 교환","mustSee":false},{"clipId":"BD3vLtWhT5A:c3:12-14","startSegmentIndex":12,"endSegmentIndex":14,"startTime":1991.2,"endTime":2019.44,"durationSeconds":28.2,"preview":"모델의 몫은 어디까지","mustSee":false},{"clipId":"BD3vLtWhT5A:c3:16-21","startSegmentIndex":16,"endSegmentIndex":21,"startTime":2024.559,"endTime":2072.639,"durationSeconds":48.1,"preview":"통신산업의 경고","mustSee":false},{"clipId":"BD3vLtWhT5A:c3:25-38","startSegmentIndex":25,"endSegmentIndex":38,"startTime":2089.359,"endTime":2228.88,"durationSeconds":139.5,"preview":"모델은 공산품인가","mustSee":true},{"clipId":"BD3vLtWhT5A:c3:49-54","startSegmentIndex":49,"endSegmentIndex":54,"startTime":2292.56,"endTime":2365.359,"durationSeconds":72.8,"preview":"앱 레이어가 승부처","mustSee":false}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:57:16.794Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":4,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 5 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["ai","distribution","big-tech","platform-shift","commodity-software","product-strategy","market-power","competition","consumer-tech","data-centers"],"normalizedKeywords":["비즈니스·전략","프로덕트","기술 트렌드"],"targetAudience":[{"who":"창업자","why":"AI 시대에도 제품 자체보다 유통과 기본값이 승부를 가르기 때문"},{"who":"프로덕트 매니저","why":"기술보다 배포·번들링·습관 형성이 채택을 좌우하는 구조를 이해할 수 있음"},{"who":"투자자","why":"거대 플랫폼과 스타트업의 AI 경쟁 구도를 해석하는 데 유용함"},{"who":"지식노동자","why":"AI에 대한 과열·반감·실제 효용을 균형 있게 볼 수 있음"}],"normalizedAudience":["창업자·스타트업","프로덕트 매니저·기획자","투자자·VC"],"summary":"이 구간은 AI가 대단한 기술이더라도 결국 많은 영역에서 '상품화된 인프라'처럼 보일 수 있으며, 승부는 모델 성능보다 배포와 기본값, 브랜드, 번들링에 의해 갈릴 수 있다는 관점을 펼친다. 모바일 전환 때 구글·메타·아마존처럼 영향이 제각각이었던 사례를 들어, AI 전환도 모든 기업을 동일하게 바꾸지 않을 수 있다고 본다.\n\n이어 애플과 구글의 AI 전략을 비교하면서, 실제 사용자에게는 모델 자체보다 어떤 제품 경험으로 노출되느냐가 더 중요하다고 말한다. 마지막으로 데이터센터 비판은 전기요금 같은 체감 이슈는 있으나 물 사용 문제는 과장됐다고 짚으며, AI 반감 역시 비용·규제·일자리 불안 같은 여러 감정이 뒤섞인 복합 현상이라고 정리한다.","insights":["AI 모델이 좋아도 배포가 약하면 시장을 못 잡는다.","상품이 상품화될수록 경쟁력은 성능보다 유통에 있다.","플랫폼 전환은 모든 incumbents를 똑같이 흔들지 않는다.","기술 논쟁은 종종 실제 비용과 과장된 공포가 섞인다.","사용자는 모델이 아니라 기본 탑재된 경험을 선택한다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c4:1-5","startSegmentIndex":1,"endSegmentIndex":5,"startTime":2400.63,"endTime":2435.04,"durationSeconds":34.4,"preview":"AI는 결국 상품","mustSee":false},{"clipId":"BD3vLtWhT5A:c4:15-25","startSegmentIndex":15,"endSegmentIndex":25,"startTime":2499.2,"endTime":2576.88,"durationSeconds":77.7,"preview":"모바일의 재현","mustSee":false},{"clipId":"BD3vLtWhT5A:c4:26-45","startSegmentIndex":26,"endSegmentIndex":45,"startTime":2576.88,"endTime":2766.56,"durationSeconds":189.7,"preview":"유통이 곧 해자","mustSee":true},{"clipId":"BD3vLtWhT5A:c4:49-56","startSegmentIndex":49,"endSegmentIndex":56,"startTime":2777.359,"endTime":2865.839,"durationSeconds":88.5,"preview":"애플·구글의 현실","mustSee":false},{"clipId":"BD3vLtWhT5A:c4:61-69","startSegmentIndex":61,"endSegmentIndex":69,"startTime":2895.04,"endTime":2987.76,"durationSeconds":92.7,"preview":"반감의 진짜 이유","mustSee":false}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:57:43.725Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":5,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 6 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["ai","artificial-intelligence","employment","future-of-work","deepfakes","generative-ai","technology-risk","social-media","career","parenting"],"normalizedKeywords":["기술 트렌드","커리어·성장","비즈니스·전략"],"targetAudience":[{"who":"AI 관심자","why":"AI가 일자리, 사회적 반발, 규제에 미치는 영향을 균형 있게 이해하는 데 도움이 됨"},{"who":"부모","why":"아이의 진로와 미래 불확실성에 대해 과도하게 불안해하지 않는 관점을 얻을 수 있음"},{"who":"학생·주니어","why":"AI 시대의 커리어 불확실성과 살아남는 직업 선택 기준을 생각해볼 수 있음"}],"normalizedAudience":["학생·주니어","지식노동자 일반"],"summary":"이 구간은 AI가 실제로 일자리와 사회를 얼마나 바꾸고 있는지에 대해, 지금 단계에서는 데이터가 너무 부족하고 합의도 없다는 점을 반복해서 강조한다. 모델 회사들이 사용량을 투명하게 공개하지 않아 논의를 뒷받침할 일상적 지표도 거의 없고, 그래서 현재의 논쟁은 경제학적 추정과 문화전쟁이 뒤섞인 '큰 흐릿한 덩어리'에 가깝다고 본다.\n\n동시에 화자는 AI에 대한 불안이 과장만은 아니라고 말한다. 딥페이크, AI 슬롭, 허위 정보, 그리고 자동화가 특정 직업군에 주는 직접적 타격은 분명히 존재하며, 과거 소셜미디어나 데이터베이스 공포처럼 기술은 늘 사람들의 삶을 망칠 수 있는 새로운 경로를 만든다고 설명한다. 다만 결론은 공포가 아니라 신중함이다: 위험을 인정하되 패닉에 빠지지 말고, 아이들의 미래와 커리어도 너무 원대한 계획보다 유연하게 바라보자는 태도다.","insights":["AI 영향 논쟁은 아직 데이터보다 추정이 앞선다.","일자리 변화는 보이지만 인과는 아직 불분명하다.","기술의 위험은 새롭지만, 공포의 패턴은 반복된다.","딥페이크와 슬롭은 '규모' 때문에 질적으로 다르다.","미래 커리어는 고정 직업보다 기술·일·보상의 교집합이다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c5:1-12","startSegmentIndex":1,"endSegmentIndex":12,"startTime":3000,"endTime":3082.8,"durationSeconds":82.8,"preview":"데이터 없는 AI논쟁","mustSee":false},{"clipId":"BD3vLtWhT5A:c5:13-21","startSegmentIndex":13,"endSegmentIndex":21,"startTime":3082.8,"endTime":3192.16,"durationSeconds":109.4,"preview":"슬롭과 반발의 구조","mustSee":false},{"clipId":"BD3vLtWhT5A:c5:22-33","startSegmentIndex":22,"endSegmentIndex":33,"startTime":3192.16,"endTime":3309.599,"durationSeconds":117.4,"preview":"아이와 진로 불안","mustSee":false},{"clipId":"BD3vLtWhT5A:c5:34-49","startSegmentIndex":34,"endSegmentIndex":49,"startTime":3309.599,"endTime":3428.72,"durationSeconds":119.1,"preview":"기술 공포의 반복","mustSee":false},{"clipId":"BD3vLtWhT5A:c5:50-62","startSegmentIndex":50,"endSegmentIndex":62,"startTime":3428.72,"endTime":3508.16,"durationSeconds":79.4,"preview":"기술은 오용된다","mustSee":false},{"clipId":"BD3vLtWhT5A:c5:63-72","startSegmentIndex":63,"endSegmentIndex":72,"startTime":3508.16,"endTime":3602.4,"durationSeconds":94.2,"preview":"일의 본질을 다시 묻다","mustSee":false}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:58:16.034Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":6,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 7 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["ai","automation","future-of-work","career-advice","productivity","technology","professional-services","business-strategy","uncertainty"],"normalizedKeywords":["커리어·성장","기술 트렌드","비즈니스·전략"],"targetAudience":[{"who":"직장인","why":"AI가 내 일을 어떻게 바꿀지와 대응 태도를 현실적으로 정리해준다"},{"who":"전문직 종사자","why":"법률·컨설팅·서비스업의 업무 구조가 어떻게 재편될지 감을 잡을 수 있다"},{"who":"예비 취업자","why":"AI 시대에 면접과 커리어를 준비할 때 어떤 태도가 유리한지 배울 수 있다"}],"normalizedAudience":["지식노동자 일반","학생·주니어"],"summary":"이 영상은 AI가 어떤 직업을 대체할지에 대한 단순한 '업무 쪼개기'식 예측을 강하게 비판하면서, 기술 변화는 늘 기존 일을 조금 더 잘하게 만드는 수준을 지나 새로운 질문과 시장을 만들어낸다고 주장한다. 음악 스트리밍, Uber, Airbnb 같은 사례를 들어 '기존 방식을 더 많이 하는 것'과 '전혀 다른 방식으로 문제를 다시 정의하는 것'의 차이를 설명한다.\n\n또한 코딩처럼 가장 자동화가 어려워 보이던 일도 급격히 바뀌는 만큼, 직업별로 몇 퍼센트가 대체될지 계산하는 방식은 신뢰하기 어렵다고 말한다. 대신 불확실성을 전제로 AI를 외면하지 말고, 실제로 써보고 익히며, 특히 전문직과 지식노동자는 AI를 도구로 흡수하는 태도가 필요하다고 조언한다.","insights":["직업은 업무 조각이 아니라 '누가 왜 고용하는가'로 봐야 한다.","기술 혁신은 기존 일을 더 많이 하게 만들다가, 새 질문을 만든다.","직무별 대체율 계산은 복잡한 일을 지나치게 단순화한 착시다.","가장 안전해 보인 직무도 가장 빨리 재편될 수 있다.","AI를 거부하기보다 직접 써보는 사람이 커리어 적응에 유리하다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c6:1-14","startSegmentIndex":1,"endSegmentIndex":14,"startTime":3601.75,"endTime":3722.4,"durationSeconds":120.7,"preview":"직업을 보는 새 기준","mustSee":true},{"clipId":"BD3vLtWhT5A:c6:15-24","startSegmentIndex":15,"endSegmentIndex":24,"startTime":3722.4,"endTime":3830.559,"durationSeconds":108.2,"preview":"대체율 계산의 함정","mustSee":false},{"clipId":"BD3vLtWhT5A:c6:25-34","startSegmentIndex":25,"endSegmentIndex":34,"startTime":3830.559,"endTime":3902.64,"durationSeconds":72.1,"preview":"인터넷식 오판 막기","mustSee":false},{"clipId":"BD3vLtWhT5A:c6:35-47","startSegmentIndex":35,"endSegmentIndex":47,"startTime":3902.64,"endTime":3980.4,"durationSeconds":77.8,"preview":"사업은 맥락이 결정","mustSee":false},{"clipId":"BD3vLtWhT5A:c6:48-64","startSegmentIndex":48,"endSegmentIndex":64,"startTime":3980.4,"endTime":4124.88,"durationSeconds":144.5,"preview":"AI 시대의 생존 태도","mustSee":true},{"clipId":"BD3vLtWhT5A:c6:65-76","startSegmentIndex":65,"endSegmentIndex":76,"startTime":4124.88,"endTime":4208.4,"durationSeconds":83.5,"preview":"실제로 쓰는 AI 용도","mustSee":false}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:58:49.263Z","keyClipsTotalSec":1154},{"videoId":"BD3vLtWhT5A","chunkIndex":7,"totalChunks":8,"title":"The most rational take on AI you’ll hear this year — Part 8 of 8","thumbnail":"https://i.ytimg.com/vi/BD3vLtWhT5A/maxresdefault.jpg","duration":4790,"uploader":"Lenny's Podcast","youtubeUrl":"https://www.youtube.com/watch?v=BD3vLtWhT5A","keywords":["artificial-intelligence","automation","productivity","voice-transcription","llm","consumer-tech","product-design","technology-trends"],"normalizedKeywords":["기술 트렌드","프로덕트","비즈니스·전략"],"targetAudience":[{"who":"지식노동자","why":"AI를 실제 업무에 어떻게 흡수할지 현실적으로 판단하는 데 유용함"},{"who":"프로덕트 매니저","why":"기능이 아니라 사용 맥락으로 AI를 포장해야 한다는 관점을 얻을 수 있음"},{"who":"창업자","why":"소비자용 AI가 왜 아직 대중적 제품으로 폭발하지 않았는지 읽을 수 있음"}],"normalizedAudience":["지식노동자 일반","프로덕트 매니저·기획자","창업자·스타트업"],"summary":"이 구간은 AI에 대한 과장된 기대보다, 실제로 사람들의 일에 스며드는 형태가 무엇인지에 초점을 맞춘다. 화자는 AI의 가치는 눈에 띄는 대형 기능보다 음성 받아쓰기처럼 '사라지는 자동화'에 있다고 보고, 챗봇처럼 빈 화면으로 남아 있는 제품은 맥락과 사용 사례로 감싸야 한다고 말한다. 동시에 소비자용 AI 앱이 아직 대중적 성공을 못 거둔 이유를 마진 구조와 제품화 문제에서 설명한다.\n\n후반부에서는 개인 취향과 삶의 태도를 통해 더 넓은 메시지를 던진다. 책을 다양하게 읽고, 고전 영화를 보고, 세상은 대체로 'it depends' 혹은 'probably going to be okay'로 접근하는 편이 낫다는 식의 느슨하지만 실용적인 세계관이 드러난다. 마지막으로 오래된 휴대폰 수집 이야기를 통해, 기술 혁신은 결국 다양한 실험이 하나의 표준으로 수렴하는 과정이었다는 점을 상기시킨다.","insights":["좋은 AI는 눈에 띄기보다 업무 속에 녹아든다.","챗봇은 기능이 아니라 사용 맥락으로 포장해야 한다.","소비자 AI의 병목은 기술보다 제품 수익화에 있다.","기술은 결국 다양한 실험 뒤 하나의 표준으로 수렴한다.","유용한 태도는 확신보다 '대체로 괜찮다'는 낙관이다."],"keyClips":[{"clipId":"BD3vLtWhT5A:c7:3-13","startSegmentIndex":3,"endSegmentIndex":13,"startTime":4211.76,"endTime":4302.719,"durationSeconds":91,"preview":"AI는 사라질수록 좋다","mustSee":true},{"clipId":"BD3vLtWhT5A:c7:27-33","startSegmentIndex":27,"endSegmentIndex":33,"startTime":4369.52,"endTime":4431.28,"durationSeconds":61.8,"preview":"표준화와 수렴의 법칙","mustSee":false},{"clipId":"BD3vLtWhT5A:c7:38-42","startSegmentIndex":38,"endSegmentIndex":42,"startTime":4466.96,"endTime":4543.76,"durationSeconds":76.8,"preview":"소비자 AI의 지연","mustSee":false},{"clipId":"BD3vLtWhT5A:c7:47-58","startSegmentIndex":47,"endSegmentIndex":58,"startTime":4566.56,"endTime":4684,"durationSeconds":117.4,"preview":"옛폰이 말하는 혁신","mustSee":false},{"clipId":"BD3vLtWhT5A:c7:63-70","startSegmentIndex":63,"endSegmentIndex":70,"startTime":4706.08,"endTime":4750.08,"durationSeconds":44,"preview":"낙관적 회의주의","mustSee":true}],"curatedSegments":[{"segmentIndex":72,"text":"um every time we have a new technology um it automates away a bunch of jobs and then that automation whether it's price elasticity and the enablement of the fact that they became automated unlocks a bunch of new jobs and so you know you go back to 1800 like 90% of us were peasants and our major concern was would like the crops going to fail because then we'll all go hungry or worse and so ever since then we've been automating jobs and creating new jobs and you can always see the job that's going to go away and you don't know the new job because it doesn't exist yet and it's like something that sounds dumb anyway like you know like railway engineer what's a railway um why would that be a thing who would care who would want to go that fast um and so we've had that process over and over again this is what any first year economics student would tell you um we've had this process over and over again since 1800 and each time you go through it you get a bunch of frictional pain and dislocation and a bunch of people lose their jobs and a bunch of towns get hollowed out and it's all it all sucks but you know when you come through on the other side we're all richer and we're not worried about the crops failing anymore and you know this is the process of the last 200 years so then the question is there some a prior reason why this would be different","startTime":1134.08,"endTime":1144.08,"durationSeconds":10,"level":"C2","overallScore":8.8,"rationale":"역사적 통찰과 고급 표현이 매우 풍부함."},{"segmentIndex":66,"text":"I think there is like tangible like my electricity bill went up which applies actually in a very small number of places objectively but it did and this is a question the water thing is weird because it's just like completely fake um and I should qualif explain what I mean here um data centers use water for cooling it's mostly closed loop but the number of data centers relative to the total amount of water use in the USA is tiny I actually went and dug into this at the Livermore lab did a study at the end of 2024 where they estimated US data center water consumption and it came out at about 0.017% of US water consumption.","startTime":2923.76,"endTime":2966.88,"durationSeconds":43,"level":"C1","overallScore":8.8,"rationale":"데이터로 주장 반박하는 설명이 풍부함."},{"segmentIndex":35,"text":"And then within that you can make sort of specific points about well how are the models going to work and do the model labs have pricing power and where's the value going to be and you know has open AAI won the whole thing or you know is anthropic got it this week and so then you can kind of get into calling those races where again it's like being in 1997 and saying well is it going to be excite or yahoo and the answer was no generally so there's a sort of a fractual point here there's like the sort a super high level that like this is going to change absolutely everything.","startTime":258.88,"endTime":286.08,"durationSeconds":27,"level":"C1","overallScore":8,"rationale":"통찰과 유용한 표현이 모두 풍부함."},{"segmentIndex":1,"text":"are like investing in buying massive comp like consultancies and PE firms talk about just what's happening there why that's happening >> well it's funny I was kind of groping for a joke last night when I wrote my newsletter and couldn't quite get to land it but as you know something like you know the joke that a machine learning scientist is a statistician who lives in San Francisco and there's something in there of like a forward deployed engineer is like an Accenture outsourced software developer who lives in San Francisco or works in San Francisco.","startTime":600.15,"endTime":628.16,"durationSeconds":28,"level":"C1","overallScore":8,"rationale":"비유로 논지를 풀며 말맛도 풍부함."},{"segmentIndex":43,"text":"What's different potentially this time just to even though your code is it's different this is everything's going to change like just like last time like the big difference obviously is uh AGI might emerge and super intelligence where that is uh could you know does the work of humans can do a lot of this stuff for us can actually replace jobs just like thoughts on that element of this transformation we're going through >> I don't know this is one of the ways I've struggled to write about AI is like certainly in like 2023 three early 24 like all the questions were questions you could have asked in like December 2022 like the questions didn't really change and the strategies didn't really change and I think the AGI question is kind of the same um I mean the thing that the observation one can make like you know we have no theory of what human intelligence is we have no theory of why these models work so well we have no theory of how much better they will get so we're all just kind of vibes forecasting as to what will happen [snorts] um and then you can have like the 2 a.m. you know doped out philosophy students talking about hey man like is this consciousness maybe we aren't conscious either we just think we are like yeah great thank you I think the one thing one can observe today is so we have no idea we don't know we can guess but we don't really know how the where this is going to end up what I think you can say today is that there's a lot of kind of redefinition of terms so I think a quote I used in my presentation late last year was an AI scientist called Larry Tesla who said AI is whatever machines can't do yet because once machines can do it.","startTime":1562.799,"endTime":1651.76,"durationSeconds":89,"level":"C1","overallScore":7.8,"rationale":"AI 전망과 개념 재정의를 깊게 다룸."},{"segmentIndex":54,"text":"You remember the idea, you remember the argu which is never a good use of time but you remember the argument of like you know people would argue about whether crypto is blockchain or whether blockchain is crypto there isn't a right answer to that let's just be sure you know it's important to understand what you mean when you say that but there isn't like a correct answer to this are we going to get to something that has human level intelligence I we don't know I don't think we have any way of answering that question maybe not you can make arguments either way meantime it does mean in the meanwhile we've [snorts] got this thing that's clearly kind of a you know completely transformative technology and maybe the serious point here is you like you don't have to believe even if like the model stopped it getting better tomorrow if this is it and we hit a brick wall tomorrow this is an incredibly useful technology that's going to change the world and get rolled out over the next 10 years so you don't have to believe in any of that stuff to believe that this is giant deal >> something that's definitely changed I had um your former boss Mark Anderson on the podcast and we didn't actually talk about this during the conversation and he brought it up before we started recording and I never got to it is he had this insight that the opportunity set for companies now is so","startTime":1734.96,"endTime":1744.96,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"불확실성과 실용적 결론을 함께 제시함."},{"segmentIndex":5,"text":"Just thoughts on that. Yeah, I mean this was his whole software is he eating the world thesis from you know 15 years ago whenever it was yeah you know the TAM it gets progressively bigger because you can address larger and larger parts of the economy and so you know if you think about the kind of the classic platform shift framing that you know mainframes are I think peak mainframe install base was something like 70 80,000 units I mean slightly fuzzy term what exactly is a main frame and what's the difference at what point does it become two main frames as one but something like that order of magnitude and then when the internet kicks off there are as I said 50 to 100 million PCs on earth maybe today there are something over a billion one to one and a half billion but obviously a lot of those are corporate it's like 7 800 million consumer PCs in the world there's about 5 and a half six billion mobile smartphones in the world which is and which is why you can have 900 million weekly IT users on JTP >> and so there was this narrative like five years ago like well we've run out of people so like this the next thing can't be in order of magnitude bigger um which was true up to a point but that was like the wrong model because clearly what's happening now is you're moving in another direction is you're just you know branching out and automating big new ways of the economy.","startTime":1821.52,"endTime":1823.36,"durationSeconds":2,"level":"C2","overallScore":8,"rationale":"비유로 시장 확장 논리를 길게 전개함."},{"segmentIndex":8,"text":"I think the kind of the other answer is you know it's back to the lump of labor fallacy and you know the last 200 years that you know each of these technologies removes a bunch of jobs, creates a bunch of new value, unlocks prosperity for all of us and that's painful as you go through it but it always creates more value.","startTime":1905.12,"endTime":1925.519,"durationSeconds":20,"level":"C1","overallScore":7.8,"rationale":"기술과 일자리의 일반 원리를 설명함."},{"segmentIndex":54,"text":"then why would you have pricing power and meanwhile if the if you need to have thousands of applications that are all different built by different people those can't all be built by the model people so it should end up looking more like cloud than it looks like Windows now that may be completely wrong and you know one of the points I make in the presentation is like imagine having this conversation about the internet in 1997 like what would you have got right and or indeed having it about mobile in 2000 you would not most of you would have missed almost all of it.","startTime":2332.48,"endTime":2365.359,"durationSeconds":33,"level":"C2","overallScore":8,"rationale":"핵심 논지를 조건문으로 촘촘히 전개함."},{"segmentIndex":69,"text":"You know in generality yes this is you know data centers are what like 5% of US energy and might grow at 1% a year for the next five years one percentage point a year but the water stuff is just nonsense and then you get into more tangible like well what is happening with this is it taking jobs away where you can watch a bunch of three-hour podcasts of econom e","startTime":2977.76,"endTime":2987.76,"durationSeconds":10,"level":"C1","overallScore":7.8,"rationale":"현상과 쟁점을 단계적으로 정리함."},{"segmentIndex":67,"text":"And certainly you know if I look at my career you know I started as an equity analyst and then I went and worked in industry and then I was a consultant like you know the days when you kind of knew what your career would was going to be or over you know there were clearly some people where you want to be an architect you want to be a software engineer you know you want to be X or Y I don't know I think you know the only kind of thinking I have here is that you have like you slowly work out there's a bunch of skills that you have and there's a bunch of like jobs that make that makes you good at and then there's a bunch of stuff that people will pay you for and you want to get at least two of those and preferably all three.","startTime":3530,"endTime":3562.48,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"경력과 일의 본질을 설명하는 통찰형."},{"segmentIndex":58,"text":"The answer I the only answer I think one can have is you know don't stick your head in the sand and say I hate all of this stuff because that gives you a great feeling of moral superiority and you can go on blue sky and shout at everybody shout at each other about how evil AI is like great I'm happy for you but that's not going to help what helps is you diving into this completely submerging yourself in it and coming out understanding what you can do with it how this changes things how can you how you can be a great hire.","startTime":4047.76,"endTime":4079.52,"durationSeconds":32,"level":"C1","overallScore":7.6,"rationale":"행동 조언과 메타 통찰이 강함."},{"segmentIndex":4,"text":"But I actually don't make spreadsheets every day. I went to a standup comedy show uh of Pete Holmes. I don't know if you know him and he made this joke that uh we want AI to do like clean the poop off the street and do all these like hard things that nobody wants to do but instead it's like oh let me help you write let me help you create imagery it's like this bohemians like no I don't want to do all these want to I don't want to do all these ugly things I want to be creative make art >> yeah well I mean there's you know variations of all of this you know it's like I don't want the AI to do the stuff I do for fun I want to do the stuff that the boring stuff that I don't do for fun >> um and you know finding that mesh I mean you joking apart.","startTime":4221.36,"endTime":4226.4,"durationSeconds":5,"level":"C1","overallScore":8,"rationale":"AI 역할에 대한 대비가 선명함"},{"segmentIndex":1,"text":"My most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile.","startTime":1.829,"endTime":5.68,"durationSeconds":4,"level":"B2","overallScore":6.8,"rationale":"비교로 핵심 관점을 제시함."},{"segmentIndex":3,"text":"Every time we have a new technology, it automates away a bunch of jobs and then that automation unlocks a bunch of new [music] jobs and you don't know the new job cuz it doesn't exist yet.","startTime":9.12,"endTime":14.799,"durationSeconds":6,"level":"B2","overallScore":6.8,"rationale":"기술 변화의 일반 원리를 설명함."},{"segmentIndex":24,"text":"Um, an interesting way of thinking about it, I did a um, a podcast last year with someone where I said, you know, I my most controversial opinion is that I think that AI is as big a deal as the internet or mobile and only as big a deal as the internet or mobile because clearly there's a bunch of people in tech who think no, this is more like the industrial revolution or something.","startTime":166.48,"endTime":185.36,"durationSeconds":19,"level":"C1","overallScore":6.8,"rationale":"비교 프레임으로 관점을 제시함."},{"segmentIndex":36,"text":"So there's a sort of very widespread of who gets it and a very wide which I think also maps this is kind of almost a separate point maps to the sort of jagged frontier question of where does this work where does it not work can you tell where it's going to work is it intuitive to know where it would work can you tell after it worked can you work out for yourself what you would do with this and all of those intersect if you're a software developer a lot of other people were like people having moment or they're not or we're in again we're in that kind of 1997 moment of okay what is this >> along those lines something you've been writing a bit about is this like unexpected investment in professional services consulting services/forward deployed engineers uh all the AI labs at least the two big ones open anthropic are like investing in buying massive I don't think it's particularly productive to say well is it 20% bigger than the internet or 100% those aren't productive conversations but it's one of those fundamental changes but then you don't know how any of it is going to work.","startTime":286.08,"endTime":303.84,"durationSeconds":18,"level":"C1","overallScore":7,"rationale":"핵심 개념과 질문 구조가 선명함."},{"segmentIndex":2,"text":"I mean, you know, joking apart, if you have any experience of professional services, like companies do not have lots of people sitting around waiting to do a build a big new project or do a big new piece of analysis or build a big new piece of technology or a new product or work out how they're going to redesign their stores or, you know, work out where the stores should be or try and work out why the churn is too high. Oh no.","startTime":628.16,"endTime":653.76,"durationSeconds":26,"level":"B2","overallScore":6.8,"rationale":"전문서비스의 현실을 일반화해 설명함."},{"segmentIndex":19,"text":"And then the third section is how does this change stuff? And one of the sort of strands I tried to pull together in the section on change is um what's the hard part of the job? Is the hard part of the job writing the code line by line?","startTime":762.32,"endTime":775.76,"durationSeconds":13,"level":"C1","overallScore":6.8,"rationale":"일의 핵심이 무엇인지 묻는 통찰형 문장."},{"segmentIndex":39,"text":"Multiply that by many, many product categories. And so what Amazon does is get you the skew, but knowing what skew you want is another job.","startTime":913.03,"endTime":919.199,"durationSeconds":6,"level":"C1","overallScore":6.8,"rationale":"무엇을 원하는지 아는 일이 핵심임을 설명."}],"generatedAt":"2026-06-22T14:59:19.664Z","keyClipsTotalSec":1154}]}