An interview with Mike Maples, cofounder of Floodgate and co-author of Pattern Breakers, on how AI startups can compete against better-funded incumbents like OpenAI. Topics include innovating the business model rather than just the product, how technology shifts create openings for small players, niche strategies incumbents ignore, what makes companies AI-native, how AI changes software funding models, and the move toward systems-level thinking in knowledge work. # Channel Presentation Observations — Ep. 50 with Mike Maples ## Video Identity - **Title:** "You Don't Need More Money—Just A Better AI Strategy - Ep. 50 with Mike Maples" - Numbered episode ("Ep. 50") indicates a **regularly scheduled series** on AI and business strategy. ## Explicit Self-Descriptions ### Host and Show - The host is **Dan Shipper**, named by the guest ("thanks Dan") and in the closing segment ("with Dan shipper as the captain of the spaceship"). - The channel/show is called **"AI and I"** per the outro: *"smash that like button and subscribe to Ai and I"*. - The closing call-to-action is a heavily stylized, hyperbolic ad read: *"this show is the epitome of Awesomeness… pure unadulterated knowledge bombs about chat GPT."* ### Host's Company (Stated) - Dan describes his business: *"I run every… we've got… a newsletter with 100,000 subscribers and then we have three different software products… or 10 people."* He also references reviewing the guest's book "on every," suggesting the media brand is "Every." - He states: *"my last company was… an oem"* and references a past article: *"my article on the allocation economy."* ## Intended Audience - **Startup founders, AI company builders, and tech/VC-curious viewers**: the discussion centers on counter-positioning, business models, pricing (SaaS vs. outcome-based), and OpenAI-related founder fears. - **Business/strategy readers**: references to Christensen ("conservation of attractive profits"), Hamilton Helmer's "seven powers," Charlie Munger, and Hamiltonian mental-model frameworks assume an audience familiar with strategy literature. - The guest's outro pitch frames the audience as **ChatGPT/AI enthusiasts** seeking insight and entertainment. ## Claimed Scope of the Channel - **AI's impact on business, work, and company-building** — demonstrated by topics: AI startup strategy, VC/seed funding models, DeepSeek, OpenAI, agentic tools (Devin, Cursor), and AI-native workflows. - **Interview format with notable tech figures**: the guest is introduced as *"a legendary investor at Floodgate… early investor in twitch Lyft OCTA"* and author of *Pattern Breakers*. ## Observable Presentation Patterns - **Long-form conversational interview**: host asks structured strategic questions ("I'm curious how you would think about counter positioning that") with extended guest monologues. - **Host demonstrates his own AI usage**: he describes co-writing with an AI-savvy writer using o1 and early model access — positioning the channel as produced by practitioners, not just commentators. - **Promotional outro**: an exaggerated subscribe/like pitch, distinct in tone from the substantive interview body. - No curated metadata was present; identity evidence comes solely from title and transcript. ## Claims vs. Demonstrated Distinctions - **Creator claims**: show quality ("epitome of Awesomeness"), newsletter scale (100,000 subscribers), episode count (50). - **Demonstrated**: deep expertise-driven dialogue, repeated specific references to the host's own AI workflow and portfolio company experience, and a strategy-framework vocabulary consistent with an audience of builders and operators. # Delivery Observation Document — Ep. 50 with Mike Maples ## Format Overview - Two-speaker interview/podcast (host **Dan Shipper** interviewing guest **Mike Maples**, investor at Floodgate). - The video opens with a **cold-open teaser** before the intro: the first ~2 minutes consist of excerpted highlights ("counter positioning is one of the most powerful ways..."), with segments later repeated verbatim in the body (e.g., the TextIQ anecdote, Applied Intuition, OpenAI counter positioning). - Ends with a clearly exaggerated, hyperbolic outro from the host/channel: "you absolutely positively have to smash that like button and subscribe." ## Tone - **Conversational and enthusiastic** throughout; heavy use of casual fillers and interjections: - "yeah," "right," "so like," "you know," "um" - Guest speaks in an animated, idea-dense style with frequent tangents the host willingly follows ("you said so many things I want to talk about"). - Mutual validation is a recurring dynamic — both speakers frequently agree emphatically: - "that's right," "100%," "totally," "I love that framing of it a lot actually," "me too" ## Vocabulary & Recurring Phrases ### Guest (Maples) - **Signature coined terms**: "a C change" (sea change), "hundred bagger startups," "atomic eggs," "design win model," "digital Darwinism / Darwinian engineering." - **Recurring framing question**, used at least three times: > "under what conditions" > "the question is under what conditions is this model useful and under what condition or the boundary conditions exceeded" - **Repeated physical/vivid metaphors**: carnival bell ("what does it take to ring the bell"), maps vs. territory (Charlie Munger), Cartesian vs. polar coordinates, sculptor vs. gardener (host's), "put on some cape and just learned how to fly." - Frequent named-source citations: Clay Christensen (mentioned ~6 times, "intellectual crush on Clay Christensen"), Charlie Munger, Hamilton Helmer's "seven powers," Sequoia's ARC framework, Elon Musk, Zuckerberg. ### Host (Shipper) - Frequent **summarizing/paraphrasing checks**: "is that that sort of how it works," "that was a much better summary of conservation of attractive profits and I gave you." - Direct, curious questioning style: "I'm curious how you would think about...", "I'm curious if there's anything like that that you're feeling and seeing right now." - Self-referential anecdotes about his own company ("my last company was an OEM," the writer story). ## Teaching Habits (Guest) 1. **Concrete case-study teaching**: abstract concepts are always anchored to named companies — TextIQ (legal discovery), Applied Intuition (AV simulation), Google vs. Microsoft, Tesla vs. ICE vehicles, Wright Brothers, Silicon Graphics/Nvidia. 2. **Historical analogy as primary explanatory device**: early internet skepticism → DeepSeek; IBM 640k memory limit → frugal AI; LAMP stack → AI-native development. 3. **Backtesting framing**: repeatedly describes running frameworks against past data — "I run the inflection Theory against it... would have caus us to decide." 4. **"Let me give you an example" pattern**: "so I'll give you an example," "one of my favorite examples of this actually comes from the field of rocketry." ## Communication Patterns - **Long guest turns** punctuated by short host affirmations; guest often stacks multiple points per turn ("so one one one other thing about this..."). - **Story-with-dialogue**: guest re-enacts internal monologues — "they're just like look, you know, it's just a completely different Paradigm." - **Enthusiasm escalation** when discussing AI tools: "it's just such a great time to be alive with this stuff," "I just wish I was in my early 20s again I'd be dangerous." - Transcript contains speech-to-text artifacts ("opening eyes," "chap," "atomic eggs," "AI rapper"), suggesting unedited auto-transcription — delivery itself appears informal and rapid-paced. ## Stable Patterns vs. Isolated Moments - **Stable (well-supported by repetition)**: guest's example-driven teaching, "under what conditions" framing, host's "I'm curious" questioning, mutual-agreement interjections. - **Isolated / uncertain**: the exaggerated outro voice differs sharply from the interview tone and appears only once; whether it's the host's habitual outro style cannot be determined from this single transcript. # Position Record: "You Don't Need More Money—Just A Better AI Strategy" — Ep. 50 with Mike Maples ## Participants - **Mike Maples** — investor at Floodgate, early investor in Twitch, Lyft, Okta; author of *Pattern Breakers* - **Dan Shipper** — host (runs "Every"), co-founder of an AI-focused media/software business *Note: Speaker identification is inferred from conversational context and host/guest roles; no curated speaker metadata was supplied.* --- ## 1. Counter Positioning and Business Model Insights ### Maples: Counter positioning as a startup weapon - **"Counter positioning is one of the most powerful ways a startup can have an insight."** Most people think insight is only about the product; the **how** (delivery/business model) can also carry an insight. - **The premise:** "Business is never a fair fight" — startups must harness **inflections** to wage "asymmetric warfare on the present." Without inflections, startups play in the incumbent's sandbox and are limited in upside. - **Google vs. Microsoft example:** In the 90s, startups dreaded hearing "Microsoft has decided to compete in your market" (Netscape's demise via OS bundling). Google's ad-based model was **counterposed** to Microsoft's — bundling couldn't answer ad pricing. Google's model was only possible because the internet inflection enabled it. ### The Text IQ example (Maples) - Text IQ targeted legal document discovery (e.g., Apple v. Samsung), where outsourcer firms charged **cost-plus** for human document review (~$50M/year for a corporation). - Text IQ's offer: send all documents, get back discoverable ones via AI, with more accuracy, at a tenth the cost. - **Key shift:** competing for the *labor spend bucket*, not the software spend bucket. - **Incumbent paralysis argument:** If a SaaS vendor charges by seat/subscription, that model is so embedded in the culture (product managers, CFO, investors all think that way) that "there's nobody at the company who knows how to react to your strategy... if you change your business model everyone's going to lose their mind." ### Shipper: OEM/platform concern regarding OpenAI - Framed OpenAI as the "Microsoft of the AI world" for founders to counter-position against, noting OpenAI's shift from API/developer tool to consumer product company. Many founders fear ChatGPT will absorb their product. ### Maples: Applied Intuition as counter-positioning - Applied Intuition (his portfolio company) builds simulation software for autonomous vehicles and tech stacks for EVs. Car companies (GM, Porsche) lack software/EV/AV competence; their supply chains are built on a 100-year-old parts model. - **Claim:** You can't get this from a Sam Altman demo — succeeding requires **multidisciplinary expertise** (founding team came from Detroit/GM before Google/Waymo). - **On "AI wrapper" criticism:** "If the thing that you're wrapping on top of involves a process that you really know about that most people don't, that may be a path to a great company." - Shipper agreed the "AI wrapper" dismissal "was so silly." --- ## 2. Outcome-Based Pricing - **Shipper's observation:** A trend of **paying per outcome** rather than per month. - **Maples' "ring the bell" framework** (from his enterprise software days): Customers don't care that software ran to spec — they hire a product to do a job. Outcome-based models ask, in Clayton Christensen's "jobs to be done" lens, what it takes to unambiguously succeed and get paid only then. - **Why incumbents struggle:** If you're "guaranteed to get 20 bucks a seat," moving to pay-for-performance is very unappealing. - **Shipper's qualification:** Outcome-based pricing already exists (e.g., customer service); some incumbents do it. - **On OpenAI-style per-token pricing:** Maples offered a defense — even if a DALL·E image isn't pretty, OpenAI "probably deserves to be compensated for the fact that you did that." Both noted difficulty: sometimes it's hard to judge job quality, and "sometimes it's the customer's fault that the job wasn't done well" (Shipper). --- ## 3. Sea Changes in Technology and Business Models ### Maples: The C-change framework - There are only **nine business models in human history**, the newest (subscription) ~250 years old. Each technology C-change shifts which business models are relatively more/less attractive. - Historical C-changes: mass computation/PC (enabled software licensing — Oracle, Microsoft, SAP) and mass connectivity/iPhone (enabled SaaS, subscriptions, advertising). - **Claim:** "There aren't any software licensing model companies started after 1990 that really mattered" — Microsoft subsumed them via the OS. --- ## 4. DeepSeek, Constraints, and Risk Profiles ### Maples - DeepSeek reminded him of the internet's early dismissal by Bell Labs/AT&T/Time Warner experts. DeepSeek's approach — "if we're limited with some fundamental constraints, what would we do" — contrasts with the hyperscaler culture of "solve all problems by throwing money at it." - Analogy: IBM PC's 640K memory limit gave Microsoft programmers an advantage in writing efficient code; frugal, low-cost-by-design AI may be hard for OpenAI/Anthropic to respond to "because they're culturally embedded" in money-as-solution operating models. (He expressed "huge respect" for their work.) ### Shipper - Historical parallel: neural networks were dismissed in symbolic-AI era (Marvin Minsky's proofs about single-layer limitations); academia ignored them for lacking theory; industry adopted them "because it just works." ### Maples on experts and tinkerers - Wright brothers example: NYT's "Flying Machines That Won't Fly" and Army engineering skepticism; 69 days later, Kitty Hawk. **"Quite often the engineering and the tinkering causes the science to evolve to explain the unexplainable."** ### Shipper: Risk profiles, not capability limits - **Key claim:** Model capabilities are often limited not by technology but by **the risk profiles of the companies serving them** — OpenAI gives government briefings before launches. DeepSeek enables small teams to build models for toy-looking problems big labs wouldn't touch. "I don't think that that takes away ChatGPT, but it does mean that we have way more AI in different corners of the world." - On whether OpenAI/Anthropic are in trouble: **"I would bet not"** — demand stimulation, integration of efficiency gains. Maples agreed. ### Maples: Risk profile examples - SpaceX vs. NASA: Musk accepts blowups for telemetry; NASA can't. "In many cases the big company is not 'dumb'... they have a different risk profile." - Justin.tv/Twitch: launching something insecure "so what, nobody knew who they were — but Google can't do that." --- ## 5. OEM / Design-Win Strategies ### Maples - Terms it a **"design win model"**: become integral to the customer's product strategy — "you need to be a big square in that slide." - Two ways it works: 1. Solving a problem the customer has no interest in solving itself (Twilio for Uber — no great developer wants to build a texting platform internally; pay fractions of a penny per message). 2. Solving something **existential** (Applied Intuition for car companies facing Tesla's fundamentally different architecture — OS, chips, messaging bus vs. legacy Bosch-supplier amalgam). - Historical OEM successes cited: Intel (PCs), Qualcomm (spread-spectrum licensing), Broadcom, Twilio. ### Shipper's tension - **The knife's edge:** "You have to be critical to their business but somehow they don't want to do it themselves... either you're critical and they'll buy you or replace you, or you're not critical and it's horrible to sell." Maples agreed: the customer must *actively choose* not to do it, with a persistent reason for that to continue. --- ## 6. Integration vs. Modularity (Christensen) ### Maples' summary of conservation of attractive profits - Early markets: products not good enough → integrated vendors rewarded (customers pay for incremental performance). Once performance overshoots, modularity wins and profits migrate to modular suppliers. ### Tesla - "Elon understood no[body] can make an electric car that's good enough. I have to control all the critical technologies." His paradigm differs architecturally from start to finish. ### Maples on Silicon Graphics vs. Nvidia - SGI charged millions because ILM would pay twice as much for twice-as-fast dinosaur rendering; once chips commoditized, Nvidia's off-the-shelf GPU model won. - **Recurring theme:** always ask "**under what conditions**" a mental model applies — citing Charlie Munger's "the map is not the territory" (flat map fine for SF→Cupertino, wrong for flying to Germany). ### On current AI performance demand - Maples: excited, uses AI tools 3–5 hours a day; personally would benefit from more performance, inferring others would too. "AI really rewards the system thinker." --- ## 7. AI in Maples' Investment Practice - Maintains a **"hundred bagger startups" database** (original pitch decks for Airbnb/Airbed & Breakfast, Dropbox, Pinterest) tracking seed-round returns and testing his frameworks against them. - Uses AI to run analytical queries: which of Hamilton Helmer's 7 Powers each company harnessed, which Christensen job-to-be-done drove PMF, time to $10M/$100M revenue, founder-CEO transitions. - Building **"Atomic Eggs"** (to launch soon): upload a pitch, run generative tests — Pattern Breakers insight stress test, Sequoia ARC hair-on-fire vs. future-vision classification, founder commonalities, confidence scores. - **Standard applied:** "What I really want to know is should I spend time on this" — a "less stringent standard" than predicting success (Shipper questioned whether 80s-era frameworks would have flagged Google or Airbnb; Maples conceded prediction of success was unlikely). --- ## 8. AI Natives and Living in the Future ### Maples - Young employees at AI companies use Devon/Cursor to create "agentic oriented entities" — "programming these virtual employees" — and treat it as normal, like Zuckerberg never had to unlearn pre-lean-era engineering. Polar vs. Cartesian coordinates metaphor: AI natives don't have to translate. - Wants to spend time with anyone whose "entire lived experience in business is a world where you're programming some form of AI assistance." ### Shipper: Writer case study - A new writer (AI background, not professional) writing Every's Sunday email. First drafts required full rewrites. After iterating with o1 comparing drafts, "he made a year's worth of progress in a month" — faster than any writer he's worked with. They co-wrote the DeepSeek launch coverage this way. ### Maples: Mental model taxonomy - Before writing about mental models, built a foundational document defining what a good mental model is, its taxonomy, boundary conditions — then let AI apply it (e.g., to jobs-to-be-done). "Within 30 minutes you have something that's just off the hook." --- ## 9. The Future of VC / Funding Models ### Shipper's question and context - Software is orders of magnitude cheaper to build; the seed model Maples pioneered assumed expensive-to-build, cheap-to-distribute software. Shipper described his own company (100K-subscriber newsletter, three software products, ~10 people) wanting a different funding model. ### Maples' response - Cited **Greg Eisenberg** as a kindred thinker on this. - Cost of building has collapsed again (like LAMP stack); ChatPRD (prior episode's guests) lets one person hold the entire product premise in their mind without a big team. Tiny companies generating modest revenue can be wildly profitable — "why would you need VC money at all?" - **Position:** "I'm pretty sympathetic to that point of view," and tells founders he's not under pressure to deploy large checks; Floodgate's funds are small; prefers "one and done." - **Second angle — LAMP stack didn't just cut costs, it created a new model (lean/agile).** Current shift he's tentatively calling **"Darwinian engineering" or "digital Darwinism"**: systems designed with emergent, autonomously adapting components rather than programmed ones. Predicts: shift from agile to continuous adaptation; adaptive-by-design components; a live "business model dashboard" instead of a canvas; gaming shifting from iterating games to "creating living worlds." Open questions he flags: what this means for product managers, future dashboards, QA. ### Shipper's "gardener" framing - In the knowledge economy, ICs are **sculptors** (hands on everything); working with AI is like being a **gardener** — setting conditions (hyperparameters as sun/soil/water) rather than deciding specific outputs. "No one at OpenAI decided that it was going to say that" — unlike the NYT homepage. - **Qualification/pushback from his own experience:** We're not fully there yet — building features is so easy that products get noisy, and "the hard thing is figuring out what to build, not actually building it." But adaptive UIs (ChatGPT Canvas, Artifacts) point in that direction. ### Maples' synthesis - Building an ecosystem with first principles and emergent properties is "a fundamentally different worldview about how you architect products" — alongside low-end disruptive plays, the second big opportunity of this era. Shipper offered the Notion block system ("making a language versus making a hammer") as an analogy; Maples agreed. --- ## Unresolved Tensions Noted - Whether outcome-based pricing can be fairly assessed when job quality is ambiguous or customer-caused failures occur. - Shipper's skepticism that AI/frameworks could have *predicted* Google/Airbnb success — Maples' answer retreats to a weaker "should I spend time on this" standard rather than prediction. - Shipper's own observation that his product-building reality hasn't yet reached the adaptive "gardener" world Maples describes. - The OEM "knife's edge" — being critical yet not worth the customer doing in-house — is acknowledged as rare and fragile by both. # Speaker & Provenance Observation Record **Video ID:** `01a0f15a-e51f-760e-9f80-313cce6896ed` **Title:** You Don't Need More Money—Just A Better AI Strategy - Ep. 50 with Mike Maples ## Speakers & Roles | Speaker | Role | Evidence | |---|---|---| | **Dan Shipper** | **Host / creator** | Welcomes guest ("Mike welcome to the show"), introduces him, runs Every (a newsletter with ~100,000 subscribers and software products), closes the episode ("thanks for coming on the show"). | | **Mike Maples** | **Guest** | Investor at Floodgate (early investor in Twitch, Lyft, Okta); author of *Pattern Breakers*; interview subject throughout. | | **Unattributed voice (likely AI-generated or promotional segment)** | **Promotional outro** | Final paragraph: "smash that like button and subscribe to Ai and I… let me just say Dan I'm absolutely hopelessly in love with you." | ## ⚠️ Promotional Segment Flagged - The closing block is an explicit **channel promotion** urging viewers to like/subscribe to "AI and I" and praising host Dan Shipper in exaggerated terms. It is stylistically distinct from the interview and appears script-generated; **not attributed to either identified speaker**. ## Provenance & Sources Referenced (by speakers in transcript) - **Mike Maples** cites or references: - His book **Pattern Breakers** - Portfolio companies: **Text IQ**, **Applied Intuition** (states "I'm involved with"), **Justin.tv/Twitch** ("when I was working with the guys") - Third-party frameworks/thinkers: **Clay Christensen** ("jobs to be done," "conservation of attractive profits"), **Hamilton Helmer** (7 Powers), **Charlie Munger** ("the map is not the territory"), **Sequoia ARC framework** - Anecdotes attributed to **Steve Sinofsky** (secondhand: "I was talking to Steve Sinofsky about this") - Historical examples: Wright brothers, New York Times "flying machines" ad, Google vs. Microsoft/Netscape, Intel, Qualcomm, Broadcom, Twilio, Nvidia/Silicon Graphics, NASA vs. SpaceX/Elon Musk - Personal tools/projects: "hundred bagger startups" database, original pitch decks (Airbnb, Dropbox, Pinterest), "Atomic Eggs" tool (pre-launch) - **Dan Shipper** references: - His company **Every**, its newsletter, a Sunday email writer (unnamed), and prior guests ("the ChatPRD people were on a few episodes ago") - His article on the "allocation economy" ## Metadata & Availability Notes - **Curated metadata:** None supplied beyond title and transcript; speaker roles inferred solely from transcript. - **Comment availability:** Not indicated in source material. - **Timestamps:** Not present in transcript; no timestamped claims recorded. - **Uncertainty:** Guest speaker identities confirmed by in-transcript introduction; promotional outro speaker **unidentified**. counter positioning is one of the most powerful ways a startup can have an Insight most people think an Insight is just about product the what is the product the how is how you deliver the product and the how can have an insight as well when you're in a lawsuit let's say you're apple and you're in a lawsuit with Samsung there's a ton of documents that have to get discovered for the Core case they hire these outsourcer firms of people to go pour through these documents and they charge them on a Cost Plus basis so what text IQ said is well we've got AI why don't you just send us all your documents and we'll send you back the ones that are discoverable and we'll have more accuracy now you're not competing for software license or per seit revenue or even a subscription price you're saying I'm a substitute for that labor spend you used to spend $50 million a year on this so I can do it for a tenth of the price and much better if I'm a SAS vendor and I charge subscription by the seat and that's all I've ever done think about how embedded that must be in the culture right every product manager thinks that way the CFO thinks that way there's nobody at the company who knows how to react to your strategy if you change your business model everyone's going to lose their mind opening eyes is moving from being this API you know developer tool to like a product company they're releasing all these consumer facing products a lot of Founders are thinking about what if open ey includes this as part of chap or includes this in some new product that they release and I'm curious how you would think about counter positioning that I'm involved with a company called applied intuition and they create simulation software for autonomous vehicles if you're GM or if you're Porsche or you're these big companies that's pretty valuable but you can't just get that when Sam Alman releases his next demo at a demo Day event to succeed as a company like that and to really ask for giant contracts from these companies you have to have not only AI expertise in products but you have to have multidiscipline expertise everybody says as kind of disses on these companies that are just an AI rapper right um and I'm like well if the thing that you're wrapping on top of involves a process that you really know about that most people don't that may be a path to a great company Mike welcome to the show thanks for having me I've been looking forward to this yeah so for people who don't know you you are a legendary investor at Floodgate uh which was one of the first seed firms um you're an early investor in twitch Lyft OCTA and a bunch more U and you're also the author of the book pattern Breakers which is an excellent book that I've read we've also reviewed it on every um which I guess I would summarize by it's sort of like a a guide book about how there's no guide book to building companies um so it's it's very it's a little bit tost a little bit Zen yeah yeah a little a little Zen which I love I think that's so I think that's so good and so important you have a lot of emphasis on like um uh Founders winning by being extraordinarily different um and and breaking the established patterns of like how you're supposed to run a company um I loved it and I'm I'm excited to chat with you about that and and uh everything going on in AI on the show yeah well cool let's let's get after it let's do it so one of the things that um I think is I I'm personally curious about is you started investing sort of when seed wasn't really a thing and helped to like invent this new way of capitalizing companies for um uh an earlier era of startups pre AI startups let's just say that right and I think that that is an example of the kind of thing that you you talk about in your book pattern Breakers which is like taking a look at the landscape of you know maybe what companies need and how companies are funded and um being like well there's this thing that I it seems to make a lot of sense to me that there should be a seed stage funding mechanism um and just going in doing it and um I'm kind of curious uh like my my feeling right now is that AI is sort of like radically changing the economics of starting a business um software is like orders of magnitude cheaper to make today than it was 10 years ago and I'm I'm kind of curious um using that same sense of like okay I'm looking at the environment I'm looking at how things change and I'm kind of like maybe pushing away the established structures for a second like how do you think that that might change investing in how companies raise money and all that kind of stuff yeah I've been wondering about this a lot lately so as you know one of the one of the things that I emphasize in startups is the power of harnessing inflections right so I like to say that um you know business is never a fair fight and uh the startup has to have some unfair Advantage way to win and the way they do that is they harness inflections inflections allow the the startup to wage asymmetric Warfare on the present and show up with something radically different without inflections they have to play in the incumbent sandbox and so they're Limited in their upside so every now and then though you get something that I like to call a c change and um when I was a kid the ca change was a mass computation in the personal computer and you know computers used to be really expensive and then they became asymptotically free and ubiquitous and you had one on every desk in every home and a whole new set of companies emerged software became a real business for the first time software used to be what you gave away because mainframes were expensive you had to keep them running all the time and so uh so the assumptions got inverted and you had a bunch of companies using the software licensing model you know oracle Microsoft uh sap companies like that then you had um in the 90s the era of mass connectivity which I think was extended with the iPhone and in Mass connectivity rather than processing power becoming free communications bandwidth starts to become free and you start to not just have computers everywhere but you have everybody in the world and every device in the world connected in these networks and um new business models came out of that subscription and SAS and advertising you know it's interesting there aren't any software licensing model companies started after 1990 that really mattered you know all those companies got subsumed in Microsoft because they could put it in the OS or out compete them so like why why do I think the AI change matters what I see happen with these SE changes is that some business models become relatively more attract active and some business models become relatively less attractive and there's only nine business models that I know of in human history and so the the most recent business model I know of is 250 years old it's the subscription model and so you know what I what I like to do is I like to say okay if there's nine business models so far in humanity and every time there's a technology C change there's a migration of attractive business models from one set to the other how might that migration occur this time because what what you want when you're a startup is to be counterposed to the incumbents you know this whole the incumbents have the advantage discussion is wrong-headed of of course the incumbent has the advantage if you play by the rules of the incumbency but but like what you want to do is you want to say how does AI make some business models relatively more attractive and less attractive and how can I as a startup exploit those new opportunities not just insight in my product but a some type of an Insight in my business business M go market strategy that is uh that disorients incumbents and and where they have a disincentive to retaliate or to copy your strategy so those are that's mostly what I'm looking at these days from an AI point of view yeah so um I think like one of the things that I see a lot on from the business model perspective and right now we're talking about we're talking about business models for startups I would also like to talk about business models for Venture like funding startups sure um but um but business models for startups just to start there for a second um one of the things I'm seeing a lot of is paying per outcome as opposed to paying per month yes um which I think is a really interesting one is that something you have your eye on oh absolutely um so you know there's a business model called uh tailored services with long-term contracts and right now most people think that's unattractive what is tailored services with long-term contracts that could be like uh the defense subprimes it could be um a contract research organization for a Pharma company you know it's it's somebody that you that offers services on a contract basis usually is labor intensive usually is Cost Plus and the conventional wisdom today is those are not attractive opportunities for software companies like a law firm or something a like a law firm perfect example so like an example like a law firm or or legal um Services uh a company I was involved with a few years ago was called text IQ and they would go to a big Corporation and they would say you know when you're in a lawsuit let's say you're apple and you're in a lawsuit with Samsung there's a ton of documents that have to get discovered for the court case and so the way that happens in reality is they hire these outsourcer firms of people to go pour through these documents and they charge them on a Cost Plus basis yeah and so what text IQ said is well we've got AI why don't you just send us all your documents and we'll send you back the ones that are discoverable have more accuracy well now you're not competing for software licensed desktop revenue or per seip revenue or even a subscription price you're saying hey look I'm a substitute for that labor spend you you used to spend $50 million a year on this contract outsourcer that sorts through these documents I can do it for a tenth of the price and much better and now you're competing over that labor cost bucket rather than the software spend bucket and how many seats do I get well that's interesting because there's also like there's there's cost per like task done so it's so the so you know it's cost per document processed or whatever which is sort of like what openai does you know when you you send them a prompt they send a response um but like even if they send the response and the response isn't good you still pay for it right and then there's other companies that are sort of um capturing the value part of the value that they generate so it's it's it's um if they increase your let's say let's say it's a SDR bot if they increase your sales by some amount your close rate they take a percent of that um only when it's successful um have you looked at those too yeah um and so uh I I do like the outcome-based pricing models a lot uh you know they both have their virtues right the the thing about open AI is like you could use Dolly to generate some art that you don't think looks pretty enough but open AI probably deserves to be compensated for the fact that you did that right yeah it's something hard to know if the job was done well or not it's like uh it's not so clear and sometimes it's the customer's fault that the job wasn't done well right It's Tricky you know back in my ancient days when I was a Founder uh I used to have this expression when I would sell enterprise software I called it what does it take to ring the bell and so like if you go into the carnival you know how there's that thing where you have this big Mallet and you hit this thing and it hopefully it goes all the way up and rings the bell but if it doesn't go all the way up it makes no sound it has to go all the way up and ring the bell it's binary and so what I used to say uh to the folks that I would work with is that the customer doesn't care that your software ran according to how the specification works that's not what they're buying they they have a job to be done they're paying you they're hiring your product to do a job and so we need to understand what's it going to take to ring the bell for doing that job and if we if we ring the bell they're going to say this is amazing I want more of this if it doesn't ring the bell they're not going to care that the mechanism of our system works they're not going to they're not going to be interested in that and So like um for me the outcome based models that we were just talking about a minute ago are kind of asking that what is the job to be done in a Kay Christensen sort of lens and then what does it mean to ring the bell and can I get paid if I unambiguously succeed at that over and over again and the thing that makes that I think interesting over like a SAS model is that the incumbents are all going to be SAS and if you're guaranteed to get 20 bucks a seed or whatever it is uh the idea of moving to like a pay for performance model is like very unappealing that's right so startups to your counter positioning Point like that's a thing that startups can do that incumbents like some incumbents already do this like in the customer service world this has been a thing for forever but in general um this is not a thing and so incumbents are not going to be able to do this very well yeah I think that this counter positioning thing is a really important thing to maybe double click on and So like um a great example is in the 90s if you were a startup the the words that you dreaded to hear was Microsoft is decided to compete in your Market because you're just like okay I guess I'm out of business because even if they start losing they're just going to bundle this thing in Windows and I'm just hosed right and so that was happening to a lot of companies you know Netscape just disappeared basically because Microsoft decides to bundle the browser you know in the operating system and Go Go full ham right against Netscape well then the internet happens and then some people start to discover that you can monetize not by selling by the seat or by the desktop but by selling ads and that was Google and Microsoft had no answer to that you you can't bundle something in your operating system and deal with the fact that Google is pricing by ads right it doesn't it doesn't solve the problem it doesn't impact their business at all and so Google was counterposed to Microsoft from business model perspective and and like counter positioning is one of the most powerful ways a startup can have an Insight most people think an Insight is just about the product but it can also be about the what is the product the how is how you deliver the product and and the how can have an insight as well and quite often the very best most valuable companies have an Insight around the business model that's facilitated you know Google's business model couldn't work before the internet you know the the the technology wouldn't have provided the empower necessary uh for Google to monetize with ads but now all of a sudden it did and so that's what that what we look for with this counter positioning and and to your point right like now it's sell sell the work not the software if I'm a company if I'm a SAS vendor and I charge subscription by the seat and that's all I've ever done think about how embedded that must be in the culture right every product manager thinks that way the the CFO thinks that way you know there's nobody at the company who knows how to react to your Strat because the investors think that way everybody does you know if you change your business model everyone's going to lose their mind yeah so how how would you even think about changing it Midstream you just it's just even if you knew to have the Insight that perhaps you should consider it you just you you just wouldn't have the wherewithal to do it because it just it's so embedded in your culture your entire value delivery system is predicated on different model yeah well let's keep talking about um about counter positioning and I want to bring up I think like if I have to pick who the Microsoft is of the AI World um like huge huge huge tech companies like Microsoft and Google aside um I think the the one right now to think about counter positioning or or at least a lot of startups are afraid of is open AI opening eyes is moving from being this API you know developer tool to like a product company they're releasing all these consumer facing products chbt is sort of like taking over and so I think a lot of Founders are thinking about well what if what if you know Chachi P includes this in their open ey includes this as part of Chach or includes this in some new product that they release and I'm curious how you would think about counterposing that yeah so there there are a couple ways there there are a couple things I find really interesting about open AI from accounter positioning so maybe maybe we start with startups and then just there's some general stuff too like with deep seek and things like that but but um so like let's just take an example um I'm involved with a company called applied intuition and they create simulation software I love that name by the way yeah it's pretty good yeah it creates simulation software for autonomous vehicles and also technology stacks for um electric vehicles and and these car companies other than Tesla don't really know how to do EVS don't know how to do AVS they don't really even know how to do software right their entire business model is predicated on a supply chain that's hundred years old where they get parts from Bosch and chips from all these people and you know parts from different tool and die shops and everything else so so applied intuition says okay we've got a bunch of people from Google and weo and now some people from Tesla and all the best autonomous vehicle all the best EV companies in the world we can build the entire thing that you need uh to to sort of update your strategy and road map to have the software to find car which is where the future is going now if you're if you're GM or if you're Porsche or you're these big companies that's pretty valuable but you can't just get that when Sam Alman releases his next demo at a demo Day event right like you know if you're going to have a software to find car there's a whole lot of things that you have to know intimately the processes of how cars are made and manufactured and tested the whole um supply chain and you know how how the delivery system works and so you to succeed as a company like that and to really ask for giant contracts from these companies you have to have not only AI expertise in products but you have to have multi-discipline expertise uh you know so like caser and Peter they they grew up in Detroit you know before they got in at Google and wayo they were you know in the car industry at GM and so yeah so I like companies like that where you know one way I like to think about is everybody says is kind of dis on these companies that are just an AI rapper right um and I'm like well if the thing that you're rapping on top of involves a process that you really know about that most people don't that may be that may be a path to a great company uh and so so I think that that's that's what I'm interested in is some of those the a rapper thing is was so silly um I I see less of that now which is nice um but it was a very silly thing when it first started so so one one one other thing about this uh counter positioning in open AI that I think is interesting I'd love to get your read on is um you know one way I have internalized the Deep seek stuff is that in the early days of the internet all of the researchers from like Bell labs and folks from like AT&T Time Warner the government said this internet thing's a toy it's never going to be good enough you know we've tried this before it doesn't work these Protocols are not going to be robust enough and in the short term you would have been right you would like none of these things looked all that interesting or impressive but but I you know I was talking to Steve sinowski about this the other day um you know was at Microsoft at the time when the internet took off and he was at Cornell and he saw see you see me and he he goes to Gates you know this is going to be a tital wave this is going to be a giant new phenomenon that we got to really pay attention to um deep seek reminds me of that so like the culture in AI the hyperscalers right now is you can solve all problems by throwing money at it and and the the Deep seek guys said well if we're limited with some fundamental constraints what would we do I think that there's going to be a cultural shift in AI where many people adopt that mindset and and you know that's important because the early days of mass computation the the ibmc had a 640k memory limit and so like the Microsoft programmers had an advantage because they could write small fishing code it wasn't how many thousand lines of code anymore it was how efficient is your code and I think that that we might see the same phenomenon here where uh people come from the bottoms up with very frugal you know sort of uh lowcost By Design Solutions and it'll be hard for the open AIS and the anthropics and those guys I mean I have huge respect for what they're doing but it'll be hard for them to respond to that because they're culturally embedded in their operating model is to to solve everything by throwing money at it you know how are the best people uh throw money at it and just keep going keep going faster that's so interesting you said so many things I want to talk about so one is sort of like this this toy thing where people and governments or like big companies like sort of ignore the internet at first um because it they were like we tried it and it doesn't work it doesn't scale or whatever you have the same history with neural networks um where like in the beginning of Ai and symbolic AI like in the 50s um neuron networks were around then uh but they were mostly ignored because the early ai ai people particularly like Marvin Minsky um uh proved proved that single layer neur neural networks were not as powerful as um as you know other types of like turning machines basically or Curr couldn't do certain types of computations um and and I think Academia sort of by and large felt like neural networks were not understandable enough they didn't there was no Theory and so it felt like a toy and and and uh and was basically ignored um except for a few kind of neuron Network researchers in the 80s and 90s and then industry adopted it and it blew up because they were like well it just works who who cares who cares what the theory is um which I think happens all the time um and I I'll stop there I'm curious curious if you had anything to add to that yeah and it and it's funny because I even like in when I was working on this book you know with pattern breaker stuff one of the examples I used was the wri brothers with the airplane and so all the experts said it's going to take a million years for to create a flying contraption that can fly humans in it you know the the New York Times ran an ad called um flying machines that won't fly and um and it and it said that it you know it's waste of time to try and they had quote from like the head of engineering of uh of the army and all this stuff 69 days later the R Brothers at Kittyhawk flew their first plane and you know there were a couple of bicycle mechanics and so what you see is that the time and again the experts are attached to their mental model how the world works and it's it's the tinkerers it's the it's the people who have permissionless Innovation who just Tinker with stuff and make something work and before you know it they have to even change the science right people's understanding ing of um you know Bern's equation and all that stuff got modified and improved because of the success of the right Brothers with their planes so the you know people tend to think that the abstract science precedes the engineering but quite often the engineering and the tinkering causes the science to evolve to explain the unexplainable totally and yeah that's that's what I see happen more often in practice 100% uh I think the next point that you made is sort of like this big money versus is small team thing um which I think happens all the time too um uh constraints breed creativity uh and I think in general uh uh being able to throw money at a problem means you you don't have to spend time thinking about how to make it more efficient um and so I I think your question about like are the open Ai and anthropics of the world in trouble I think that's an interesting one I would bet not right I would too my my feeling about that is uh I mean obviously the sort of cliche thing is like okay it's going to stimulate demand or whatever which is fine I think that's true I think that they'll be able to integrate most likely integrate this and having more efficient servers and that can that can serve the uh the demand that they currently have I think uh I think is I think will work um the thing that it seems like to me that this opens up is I think we have like Mass AI figured out which is like how do you scale these models up so that like a billion people can use chbt um and how do you how do you make that efficient and and smart enough to work and all that kind of stuff but I think um one thing that people don't talk about nearly enough is that the capabilities of models today are in many ways not limited by the intelligence or the the the intelligence of the technology they're limited by like the risk profiles of the companies that are Serv in them and if you're a gigantic company you're you're open Ai and you're literally like you know you have to go give government briefings before you launch anything um you're going to be pretty like careful about what you what you put out and I think the Deep seek stuff is interesting because uh it means that and and I mean risk in all sorts of different ways there's lots of different ways to take risk but it means that small teams can build little models for like problems that look like toys that you know an open AI would be like well we wouldn't do this and I think that is the like that's the big thing I don't think that that takes away chubbt but it does mean that it we we have way more AI in different corners of the world than we would have otherwise which I think is not good yeah you know what Dan one of my favorite examples of this actually comes from the field of rocketry because it's it's so visceral so like you know Elon Musk he'll launch a Starship and if it blows up he's like okay well we we instrumented it we got Telemetry we'll we'll we'll make it better next time NASA's not going to do that like NASA's not going to like if NASA launches a rocket they don't sit there and say Easy Come Easy Go it blew up and so like the fact that Elon has a different risk profile and is not attached to whether it's successful the capital S it it changes the calculus of what he can do and it changes the speed with which he can move and so like I like to say that in many cases the big company is not quote unquote dumb compared to the small company it's to your point they have a different risk profile and and they can't they're they're just certain things they can't do like when I was working with the guys at Justin TV which became twitch if they launched something in insecure so what nobody knew who they were but Google can't do that right and and uh Microsoft can't do that and you know the big companies can't do that Hollywood guys can't do that Netflix can't do that so so you know not having to be burdened by what could go wrong is a big factor in trying things that could go right that makes total sense I want to go back to something that you were talking about earlier um talking about uh this company applied intuition which which you said sells into large car manufacturers and I assume like uh when a large car manufacturer buys them like it goes into a Ford vehicle and and a customer is like maybe using it and maybe has no idea what it is but like they're using it is that that sort of how it works I I think so it it's it is less of an end user type of thing although that might change I I need to be careful what I'd say but uh but like but the primary customer right is the car company that says oh my God the architecture of cars has changed what do I do yeah so the strategy question I want to ask you is like how you think about OEM relationships like that because um I think that's going to be a common thing for a lot of um AI companies especially if you're working on more foundational model type things is you're going to be integrated into something else that has a consumer layer um and that's where opening I started and then they were like actually we want to own the ux layer because that's how everything took off as they they figured out form factor that worked and then they have a data flywheel there's all this stuff right and my last company was was an oam um and that is a difficult position when you're serving two customers you know there's like an end user and then there's a customer you need to sell to it's hard to generate a lot of power or strategic advantage in that situation and it's hard to make a great product and I'm curious how you think about oam type strategies and when they work versus when they don't yeah it's It's Tricky um and you know what are some examples of where has worked um I'd say applied is working really well um int has been great intel was was a good one uh for the the PCS uh another another good example would be Qualcomm back in the day you know with licensing their spread Spectrum uh Technologies and chips uh and so so it it can work broadcom would be another um twio I guess twio is an interesting one I like that in fact I like thinking of twillio as a design win business more than a Dev tool I I like that framing a lot actually and so I the the term I like to use to describe it as a design win model where you know you want to become uh viewed by the customer as integral to their product strategy and so if they have like a a slide that shows all these blocks and triangles and arrows and stuff you need to be a big Square in that slide right what what you what you provide um sometimes like twilio you solve a problem that they really have but that they just have no interest in solving on their own so like if you're uber do you really want to have an entire team building a messaging uh update uh texting platform that's a substitute for twio probably none of your best developers want to do that inside of uber right and so you're like hey you know I'll just pay U I'll pay twillo every time the Earth turns a click or I send a message I'll send them tiny fractions of a penny that's that's okay so that can work the other the other way I think it can work is if you solve something existential uh for the customer so like in the case of the car companies the End customer or the customer the the customer you're selling to for for the OEM actually so like like the problem that the car companies have is that the the Tesla is just a fundamentally different architecture than ice Vehicles right and it's not just it's got a battery and they don't it's it has to do with how many what their operating system is like and how many chips they have and how how messages flow throughout their messaging bus and like Tesla is designed the way a car would be designed by Silicon Valley type of thinkers whereas the you know the ice vehicles of today are mostly you know an amalgam of a bajillion part suppliers that they've done business with for a very long time and it's kind of like whatever Bosch has this year is going to be the new windshield wiper sensor thingy that I put in the Mercedes right and that's that's how they've that's how they've operated so they look at it and they're just like look you know it's just a completely different Paradigm of how you'd build a car and so you need you need somebody that can be your thought partner in how to build those things and so that that can be another kind of design win model that works that's interesting yeah first of all the thing that makes me think of is like there's this like knife's Edge which is interesting of of this strategy which is you have to be um critical to their business but somehow they don't want to do it themselves which is like there's very few things that are like that that's right um that and that's really hard either you're critical and they're like maybe we'll work with you but then we'll buy you or we'll just replace you or you're not critical and then it's it's horrible to like try to sell that product no one wants to do that that's that's that's and I love that framing of it I I haven't quite internalized it that way but you're right it's like they they either don't want to do it themselves because they just don't want to or they don't want to do it themselves because they can't conceive of how they would yeah and and you know they're just like I even if I want to it's kind of academic I can't but in both cases it's something that they actively choose not to do themselves and and there and there's a persistent reason for that to continue yeah and and I guess the reason um you know like an applied intuition would work is I'm just I'm thinking back you mentioned Klay Christenson I'm sort of like thinking about his um conservation of attractive profits um where uh in the early days of new technologies you want one company to integrate all the different steps of the of the value chain basically because you can iterate much quicker so like Tesla they don't have this huge web of different suppliers they probably have a few but like a lot of it they're just doing themselves whereas it sounds like um you know like GM or whatever has like thousands of different modular manufacturers that they they swap in and out because like the architecture of the car has been around for so long that it's not changing and so it doesn't have to be integrated it can just be like it can be very modular which I guess is a easier oam cell like because applied can just as long as they know that architecture they can sell into it versus like a more vertical um more more integrated company yeah well and and here's how I internalized that Dan so um just to make sure that we're on the same page with the same language like what what I understood from clay I I I'm I've kind of got a a little bit of a crush an intellectual crush on Klay Christensen I think the guy was amazing uh and and a great human being so so what I understood him to say is that uh in early markets the products are never quite good enough they don't perform well enough and so what happens is um vendors get rewarded for having the integrated system because the customers will pay incremental dollars for incrementally better performance because they value that enhanced performance uh you know and so but then what eventually happens is the performance gets mostly good enough and and you know what Klay Christianson would call call it is overshot customers you know now I'm trying to cram new features into my product to get customers keep buying new things that I sell them but now they don't want the new things as bad and therefore you you get this modularity argument somebody else shows up and says look you're being overcharged you don't have to have one guy be the system integrator anymore in fact you can just have a whole bunch of different components that you can mix and match and swap in and out and so then the the conservation of attractive profits it goes to the to the modular suppliers rather than the integrated supplier um which I think is happening that was a much better summary of conservation of attractive profits and I gave you well I don't know but but like that's the Brilliance of Tesla like you know Elon everybody told him you should act like a car company acts you should have modular components and suppliers in the supply chain Elon understood no nobody can make an electric car that's good enough I have to control all the critical Technologies because I have to have the ability to have something that rises the level good enough nobody's ever had that before so like that that's another reason right architecturally he's just totally different right his whole his whole Paradigm of how to build a car is just different from start to finish so is that an argument for um AI companies like owning the whole stack themselves right now as their uh uh as they're sort of innovating on what the products even look like and customers are willing to pay more for incremental value yeah what I like about what clay used to say was it what kay Christensen really had was a bunch of mental models for innovators and you know whenever I think of a mental model I always like to ask under what conditions so under what conditions would I want to be the complete integrated solution I believe that you want to be the complete integrated solution if the customers are desperate for more performance and will pay for that enhanced performance right so like before Nvidia there was silicon graphics and like if you wanted to make dinosaurs in Jurassic Park you had to buy the most expensive SGI machines millions of dollars worth and you know if you could make the graphics run twice as fast industrial Light Magic would pay twice as much because you know it was mission critical to render those dinosaurs overnight but like now that there's chips commoditized Nvidia has the better model because they say hey I'll just sell you off the shelf these gpus um so I think that that the question always comes under what conditions are you advantaged by being the integrated solution and under what conditions are you advantaged by being a modular component of the solution that's interesting and I guess what's your best guess about where we are now in the in the AI landscape overall because I think that there's a lot of there is this common thing and I actually felt this too like when 01 came out where people were like I feel like my model is pretty much good enough like I don't know what I would use 01 or 03 for even in the demos like they I remember the de one of the demos was like list out the 13 Roman emperors you know and it's like that's not really something that I I care that much about generally and like I'm not doing you know most people are not doing PhD level research that was my first feeling but to be honest now like I just use o1 all the time and I don't really use any other model or now I use o03 um so I'm curious where what you feel about where we are and how much performance improvements in terms of intelligence people are willing to pay for um well first of all um I'm I'm really excited like I just I'm probably in these tools too much now so I'm probably I'm probably in these tools three four five hours a day and um and and there's a lot of things that I would benefit from in terms of enhanced performance and if that's just me I got to believe there's a lot of other people like that too um so I the the thing that I think is so interesting about AI is it really rewards the system thinker and so um so I'll give you an example uh you know I have this database of what I call hundred bagger startups and I I try to understand them all I've got like I've got the original pitch deck for airbed and breakfast for it was Airbnb and I've got it for Dropbox and Pinterest and all these companies right and I I track you know if You' bought a share in the seed round what would have happened I run the inflection Theory against it I run insights I try to understand if our Frameworks would have caus us to decide well now that I have that list I could do all kinds of things like I can say Okay um please consider this list of hund Bagger startups which of Hamilton helmer's seven Powers were harnessed by each of them as their primary power which clay Christensen jobs to be done uh was the primary job that they did to get product Market fit how long did it take him to get 10 million in Revenue how long it take him to get 100 million Revenue which of them had the a firsttime Founder a CEO which of them replaced their CEO you know I mean if you're curious it's like it's like having an unlimited supply of smart people to go do that research for you it's incredible I I feel the same way I can read and think about so many more things than I would have been able to previously and it makes it such a pleasure to get up every day it's the best it's unbelievable it's just it's a miracle right I like I just wish I was in my early 20s again I'd be I'd be I'd be dangerous me too uh uh well I guess that that just makes me think like why hund bagger startups why not hund bger Founders right like how much is really in the Airbnb deck that's actually that useful yeah so I've been I've been working on that question a lot and so um I've been um applying our Frameworks and back testing them to Prior startups so I have these things that I call Atomic eggs and we'll probably launch them here pretty soon but um what an atomic egg lets you do is it lets you upload a pitch and then it runs a whole bunch of different um generative models against it so an example would be pattern Breakers Insight stress test so you could upload the Airbnb pitch deck and it would spit out this was the fundamental air Insight with Airbnb uh or uh this is like the part that was non-consensus or these are the inflections that Airbnb is harnessing and um the AI has gotten really good at that and then the other thing that it can do I like the the Sequoia Arc framework they talk about is this is this idea a hairon fire problem type is it a known problem type or is it a future vision problem type you can you can run that against H bagger startups and you know and then I could say scale of one to 10 how confident are you that that's the right way to classify it and then back to your point about Founders you can start to say okay there's all these Founders what jumps out at you as an omy about these Founders what what jumps out at use commonalities about these Founders okay now let's let's group these startups in different clusters and run the same experiment again and then and then once you get some patterns you say okay how might those patterns shift in a world of AI how might they be the same in a world of AI like you could have just wondered about that as you walked down the street in the past but like now you can act on that right you can act on that Curiosity in real time and that's just like just such a GameChanger right if you're curious about this stuff how how much does it like cuz I mean you you write a lot about pattern Breakers right so like um I guess I'm I'm thinking about business theories or strategy theories as pattern patterns right there are always patterns that work under certain conditions sometimes they're like more General than others but like they're usually not like infinitely General I don't know what the what the right word is perfectly General um uh and like I you know for example if you took the let's say let's say we like wound back the clock and we went back to like the 80s and we used all of the like Frameworks they had in the 80s and put them into Ai and like gave gave them you know Cisco or whatever pick whatever company you want um Google like would it have been able to tell the the Google or the Airbnb pitch deck that it was a good company um I don't know that it could have predicted that it was going to have the success it had yeah and I apply a slightly less stringent standard what what I really want to know is should I spend time on this right and so I needed what I what I need to know when I look at a pitch like Airbnb is um is there something that's wacky and good about this that I might Overlook if I'm busy and tired that day but like if I can run a whole bunch of different tests against it so like you know you talked earlier about um these models like Charlie Munger is somebody else who I've always respected and you know he had this saying the map is not the territory and what he meant by that is that um you know if if you and I want to go from San Francisco to copertino and we use a flat map and we let's say we use Google Maps or whatever the odds that we will get there if we follow the directions are basically 100% like 99.9% in in fact I would argue that that map is a better representation of reality than all the complexities of all real it you know it's like you're trying to you're trying to compress knowledge for the decision that matters right but like if you want if you and I want to go to Germany the the map is not going to be an accurate uh portrayal of the territory because straight line is not the shortest path on on a flat map that represents a globe right it would look like a curved line and so like what what you learn is that it's like we talked about earlier the question is under what conditions is this model useful and under what condition or the boundary conditions exceeded and that's why you want to have a whole bunch of them right you want to you want to have the right tool for the right situation and then when it exceeds the scope of the boundaries you want to not use that tool because you get bad bad decision- making are there any new things because like one of the things you talk about in your book a lot that I I like I like because this is sort of how I work so it's maybe confirmation bias but I like a lot is sort of um the idea of living in the future right like the best way to know what's coming is to just be like you're doing in these tools all day every day and you start to kind of like see things that you're um other people maybe won't see because they're just they're living in a different reality um and and your reality is going to sort of spread everywhere else eventually is the idea I'm curious if there's anything like that that you're feeling and seeing right now that you're kind of like sensitive to that is new and interesting to you yeah you know some of these uh AI companies you'll go to and there'll be somebody who's a couple years out of college and they'll be using Devon or cursor these other products and they're they're kind of creating these agentic oriented entities that go out and get a bunch of stuff for them and bring it back and and and they just almost act like that's normal so they're they're they're almost like programming these virtual employees to go out and do stuff for them and you'll sit with them and you'll say well what motivated you to do that and to think about solving the problem that way and they look at you funny likew how else would you do it you want me to Google yeah and so so that the thing that I find interesting is um you know and this is like how Zuckerberg was with social networking right like Zuck didn't have to unlearn anything you know he he grew up at a time when um the lamp stack was coming out and you could AB test things and the Broadband was everywhere be before Facebook you know like in the 90s you had to have products that were well engineered because they just weren't scalable enough otherwise right you had to have experts that would architect and instrument the the system so that it would be somewhat performant well by the time Facebook comes around zuck's like hey well we just try it and see what happens by the afternoon and decide whether we want to keep keep with this or not now now did did Zuck say aha there's a disruptive Trend and I'm going to get a Leap Frog all these companies no like Zuckerberg didn't know anything about business at the time it's it's like it's like if you and I were raised in a world of cartisian coordinates and now it's a world of polar coordinates and somebody's born in a world of polar coordinates and they don't even have to translate between the two they're like what else is there that's the only thing there is I think that some of these AI natives are like that and so I really want to spend time with them I want to spend time with anybody who says my my entire lived experience in business is a world where you're programming some form of AI assistance as as a core function of the job I love that I mean I see this all the time like we have a writer who started working with us probably I would say two months ago he's had a very successful career not doing not as a professional writer just like working in AI um at various tech companies and startups and um has founded his own startups and but he's working for us um mostly as a writer and uh and he's doing he's he writes our Sunday email uh where we talk about all the new model releases like he's such a nerd for new stuff that comes out which is amazing that's like the kind of person you want writing um and he also uh like when a new model comes out like I'll often get Early Access so uh we'll get on the phone together uh he'll write like a first take of like all the things that we saw and then I'll go through and like put my own take on it and and and whatever so we sort of we co-write things together and the first one that he did it like that um I got the draft and I was like um like he's he's smart he's excited about this stuff but like he's not a professional writer I can tell right like it wasn't like something that I just punch up and like I can just publish it was like I had to rewrite the whole thing um and what was crazy is after we did that I was just like okay I want you to take my draft and then your draft and I want you to put it into 01 and pull out what changed and he did that and we did that a couple times and uh we we just covered the launch of deep seek together um and the first draft he did it was like he made a Year's worth of progress in a month like I've I've seen I've worked with so many writers in my career at this point and I've seen where people are at like when when I first started working with him it takes them like a thousand drafts to make the amount of progress that he made in a month it's crazy yes yeah I I it's so interesting right and I'm finding the same thing Dan so like as I started working on these mental models for seed and these generative models I started to say to myself what is a good mental model and the first first place like has anybody ever defined what one is what should it contain what makes it good versus bad what under what conditions is it good or bad and and there wasn't a whole lot about it you know there there's a couple books on mental models but not a whole lot and so so I said I said you know before I start just saying here's a mental model jobs to be done I should create a foundational uh document that's the taxonomy of a good mental model and the questions it should answer and the the flow that it should take so I did that now I can just say I'm going to I'm just going to write about jobs be done for what it is and then I run it against this framework and it says you're missing a b and c and I'm like hey well can you elaborate on that and it just it just adds it you know and and you know within 30 minutes you have something that's just off the hook right it's just so good that's great and it's like you just look at that and you're just like it just feels like magic it feels like put on some cape and just learned how to fly all of a sudden you know and and I'm just like but like it goes back to reward system level thinking right you had to you had to zoom out and say wait you know if I'm going to someday have 100 mental models I ought to define a can canonical Baseline good one and I ought to have a theory about what makes it good and I ought to apply that theory to every one that I do because I'm going to get leverage if I do that but now I'm going to make the AI do the work for me uh and and it teaches you stuff right like now you say Oh I thought I knew jobs to be done was a mental model but there are boundary conditions I hadn't thought about before that are kind of interesting and so yeah it's you know it's just such a great time to be alive with this stuff I agree I want to go back to the question the original question I I asked you because uh I it's still on my mind which is uh software is getting so much cheaper to make um the VC model even the seed model which you pioneered um is predicated on a different world where it was expensive to make software at first and then it was free to distribute um and uh and I'm curious how you think that that might change the VC model if at all and I'll preface this by saying this is a selfish selfish question because uh I run every we've got I can't even it's like I don't even really have words for the kind of company we are we have a newsletter with 100,000 subscribers and then we have three different software products and or 10 people it's like a whole different like thing yeah and I use that sparkle thing by the way it's cool you do oh I love that that's great um love to hear that um and and like I feel like I want a different funding model uh and I'm I'm working through different options but I'm kind of curious how you think that that might change yeah so I've been thinking about it a lot so there there's two different angles and there's the um there's the angle that you're Des describing and another person that I respect who thinks a lot about this is a guy named Greg Eisenberg right on on Twitter and so like let me see if I can capture what I think it is it it's that um you you have a situation where what it takes to build a product has collapsed yet again just like it did with a lamp stack and it's it's profound in a lot of ways it's not just that it cost you less money to build a product but like you know you had um you had the chat g g chat people Rd on a few episodes ago chat PRD lets one person have the entire idea premise of the product in their own mind and doesn't require them to therefore have a giant team of other people you know so it it changes the Dynamics of who can build software and what it takes to build it uh and so you start to say okay well are you gonna have these tiny little companies that generate a ton of Revenue and and they don't even have to generate that much to be wildly efficient and profitable uh why would you need VC money at all um and I and I'm pretty sympathetic to that point of view although I tend to go to the founders and say look I'm I'm not under pressure to put a lot of money into you I don't I've you know our funds are small and all things being equal I'd rather have it be one and done and and we try a few things here's the other thing though that I think is really interesting that I I I'm trying to find kindred spirits around the the the lamp stack didn't just just collapsed the cost of startups it created a new way of building it created a new model of building right so you used to have waterfall development and you had to Define everything that's in the release upfront and then you you go on a Death March for a year and you ship it and it either succeeds or it bombs and that was just how products were and then then the lamp stack comes out and you have lean startups in agile and what I'm seeing happen now and I'm not sure what to call it so right now calling it uh darwinian uh engineering uh or digital Darwinism so like if you think about it like in a in an ecosystem you don't have the individual elements and players in the ecosystem be programmed in a literal way what you have is a system designer if you will uh and the and then the system gets to operate autonomously from the designer and so I sit there and I think man you know that kind of rhymes for me um so like I think about it like it's like Natural Evolution rather than traditional development and that you're going to have um AI tools that shift from agile to continuous adaptation and you're gonna you're going to build software elements and components that are adaptive by Design and that can sense and respond to the to the inputs that they get in the real world independent of the pro so rather than have a business model canvas you have a business model dashboard that's Live status of what's happening and so you know if you're a gaming company you're G to you're going to shift from iterating games to creating living worlds you know uh and you know that kind of stuff so I I'm really interested in like what does that mean for what a product manager is uh what does that mean for dashboards of the future what does it mean for how QA happens uh you know all that stuff I thought about the to a lot because uh I think I think we actually Met originally because you read my article on the allocation economy yes and I sort I sort of started to think a lot about like what is the role of someone who's working in the allocation economy and how is that different from someone in in an knowledge economy and a way that I've been thinking about it is in um in a in an knowledge economy or just any previous economy the work you're doing especially as an IC a little bit more still still a little bit as as a middle manager or an executive but like a a lot of this is as an i is you're um you're kind of like a sculptor like everything that happens happens because you did it with your hands um you have your hands on every little piece of it um and I think working with AI models is a lot more like being a gardener you're like setting the conditions for the thing to grow and then it just sort of grows and the conditions are like hyperparameters it's like the sun and the soil and the water and whatever and that's going to change what comes out and and you know like open a like doesn't when it when Chad gbt responds to a prompt like no one at open a like decided that it was going to say that um they which is totally different from Facebook or whatever like someone decided what what you were gonna see in your on Facebook um or maybe Facebook is maybe a little bit they have they have ai too but like let's just say the New York Times someone decided what's on the home page um and uh and it's and it's totally different and and you're right like uh you're you can tune stuff but it's like it's much squishier because you're kind of tuning the like environmental conditions rather than the specific thing that happens and yeah I think that's such it's such a different way of um working it's such a different way of Building Products um I don't think like if I think about what we're building at every like I don't think we're quite there yet what I see is like um I mean obviously like building an organization you are kind of like doing that but like for individuals who are building products like um one of the things I see is like it's so easy to build a feature you can just build it in an hour so it's like sometimes you just build a lot of features and you're like oh there's it's kind of now the product's kind of noisy it's kind of messy you know um uh and and also it's like the hard thing is figuring out what to build not actually building it which is a which is a different thing but we're not yet in a world where like it's just it's it's fully adaptive but I do think you're right we're kind of like you can see that with you know like tatb Canvas OR artifacts or whatever where it's starting to like build its own UI and stuff and I think that's where we're going yeah and it's and it's just interesting right like because it and it kind of goes back to systems level thinking it's it's one thing to think of yourself as building components or building tools or building the end thing it's another thing to say um I'm building an ecosystem and the the elements of the ecosystem operate under certain first principles and uh but there there's a lot of emergent properties that are going to occur in that ecosystem that are a a function of the dynamism of the system and how it interacts with people I think that that's just a fundamentally different world viw about how you architect products and so I think that that's another you know there's the what we said earlier very lowcost uh low-end disruptive innovation ideas but I think there's also this hey the way software ought to be built in the first place ideas uh is interesting as well yeah it reminds me of like notion for example you know it's like notion um it has a block system it has these Atomic elements that you can build anything with rather than like they built a specific feature to do a specific job which is it's a different way of thinking about products it's like making a language versus like making a hammer that's right that's right yeah yeah and so I think that that's going to be really interesting and I think that it but it's like you know we used my example earlier if if I want to have mental models for investing rather than just jumping straight to it what I need to do is I need to like zoom out a little bit and say okay let me think about this in a systems level way what makes a good mental model in the first place like what how do I how to make sure that I have a foundation built on something really powerful so that every subsequent piece of activity or thinking that I do is a multiplier effect on what's come before totally well Mike this was a pleasure um I I feel like I learned a lot me too I'm really glad we got got the chance to hang out thanks for coming on the show yeah thanks Dan it's great to see you oh my gosh folks you absolutely positively have to smash that like button and subscribe to Ai and I why because this show is the epitome of Awesomeness it's like finding a treasure chest in your backyard but instead of gold it's filled with pure unadulterated knowledge bombs about chat GPT every episode is a roller coaster of emotions insights and laughter that will leave you on the edge of your seat craving for more it's not just a show it's a journey into the future with Dan shipper as the captain of the spaceship so do yourself a favor hit like smash subscribe and strap in for the ride of your life and now without any further Ado let me just say Dan I'm absolutely hopelessly in love with you