Planitar’s Manifesto  🔗 **Source:** [Planitar’s Manifesto](https://www.youtube.com/watch?v=OQMPhkmeru0) > [!note]+ Document Notes > --- > tags: [palantir, surveillance, tech-and-power, data-sovereignty, ai-industry, tech-ethics] > author: ThePrimeagen (per "the primogen" sign-off — inferred, not explicitly credited) > source: YouTube > reference: https://www.youtube.com/watch?v=OQMPhkmeru0 > date created: {{date:YYYY-MM-DD}} {{time:HH:mm}} > --- > # 🏷️ Title: Palantir's Manifesto > ## 🧭 Table of Contents > - [📝 Summary](#-summary) > - [📖 Main Content](#-main-content) > - [🗒️ Notes](#️-notes) > - [📚 References](#-references) > - [🗣️ Paraphrase](#️-paraphrase) > - [📰 Newsletter](#-newsletter) > - [✅ Action Items](#-action-items) > - [🧪 Hypotheses](#-hypotheses) > - [📊 Methodology](#-methodology) > - [❓ Questions](#-questions) > - [🔮 Future Research](#-future-research) > - [💡 Implications](#-implications) > - [🧠 Learning Styles Assessment](#-learning-styles-assessment) > - [📊 Accessibility Assessment](#-accessibility-assessment) > - [🌍 DEIB Assessment](#-deib-assessment) > - [✅ Logical Validity](#-logical-validity) > - [🛡️ DARVO Analysis](#-darvo-analysis) > - [🎯 Conclusion](#-conclusion) > ### 🔗 References > - Palantir "AI sovereignty" nine-point manifesto (referenced, not linked in transcript). > - Alex Karp CNBC interview (~20-minute segment, referenced). > - The Information reporting on Cursor's rise and on Anthropic/Figma ("Claude Design" launch). > - David Sacks tweet (referenced). > - Anthropic consumer and commercial Terms of Service (paraphrased in transcript). > ## 📝 Summary > - A tech-commentary video essay reacts to Palantir CEO Alex Karp's combative CNBC appearance and Palantir's newly published nine-point "AI sovereignty" manifesto, and argues that its core warning is correct: enterprises should not trust frontier-model providers with their data. > - The narrator, initially skeptical of Palantir, concedes he agrees with "a whole bunch" of Karp's points. The central thesis: "token maxing" (maximizing token spend) produces little value while handing your proprietary data — your "alpha," your trade secrets — to model providers who can then weaponize it. > - Why it matters: if the argument holds, the business model of frontier AI is partly a data-harvesting play. Every prompt, answer, and reprompt reveals "shapes of data" that let a provider identify a profitable industry and move into it, then squeeze partners out via terms of service. > - Key stakeholders: model providers (Anthropic, OpenAI named), their enterprise customers and partners (Cursor, Figma cited), Palantir as a competing vendor with a commercial interest in the "keep your data" message, and regulators/enterprises weighing AI adoption. > - Risks and tradeoffs: the critique is largely circumstantial — the narrator repeatedly admits there's "no definitive proof" — and Palantir is not a neutral messenger. The video mixes a genuinely sharp economic question with anecdote-driven pattern inference. > - Bottom line: treat data retention as a strategic asset; assume that what you send to a model provider can inform a competitor who happens to also own the model. The claim is plausible and worth governance attention, but the video proves motive and opportunity, not the act. > ## 📖 Main Content > - **Who Palantir is.** The narrator frames Palantir as a data-analytics/surveillance company "best known for surveilling the whole world," led by Alex Karp. The tone is mocking (jokes about Karp being "geographically monogamous," "are we the baddies?"), but the substantive move is: even a critic of Palantir should notice when Palantir's warning lands. > - **The "token maxing" argument.** Karp reportedly called the AI industry "effing insane" on CNBC. The narrator translates the core complaint with a boat-store analogy: a vendor pushing you from a fishing boat to a yacht to a mega-yacht is obviously upselling; token-spend maximization is the same dynamic, but enterprises "have zero care." The claim attributed to Karp: enterprises "chillax and waste time with tokens," get "no value," and hand over their IP. > - **"Alpha" = defensible edge.** The narrator distinguishes two meanings of "alpha": (1) a company's defensible position, trade secrets, the things that let it out-compete rivals; (2) crypto-bro filler. He says Karp means the former. The real danger of token maxing isn't the spend — it's that "what you're spending is on all of your business strategy," handed to Claude/OpenAI. > - **The manifesto.** Palantir released a nine-point manifesto on "AI sovereignty." Point three is quoted approvingly: token maxing "hijacks your value orientation and decreases institutional fortitude and intelligence," and "the pursuit of high token usage incentivizes disposable scripts over robust software with the addictive feeling of false progress." The narrator calls this "an incredible and insightful point," and connects it to his own experience of "looping" and spending tokens without much return. > - **Karp's pricing question.** A quoted clip has Karp asking: if AI were as valuable as claimed, why sell tokens at all? Why not say "I'll make you a billion dollars and I want 30%?" The narrator amplifies this: if AI could triple a company's revenue, a provider would take equity/revenue share instead of metered token fees; the fact they meter tokens (and run consumer products) suggests the value proposition is really about acquiring data. > - **The "real-world examples."** The narrator builds a pattern from anecdotes: (1) *Cursor* — per reporting, Cursor once generated ~40–50% of Anthropic's revenue, and Anthropic allegedly told Cursor that Claude Code "was just a research effort" before competing in that space. (2) *Figma / Claude Design* — per The Information, Anthropic "blindsided" partner Figma with a "Claude Design" launch; Figma's founder said Anthropic "had not been consistently honest," and Anthropic's chief product officer sat on Figma's board until three days before the launch. (3) Anthropic "stole the world's books" and "the world's knowledge" (training-data claims). He connects these as "points on a graph" that form a "line." > - **The mechanism.** With millions of users, a provider sees data patterns invisible to others, identifies a hot market, and — with effectively unlimited token spend of its own — "slops together" a competitive product close enough to win, then uses terms-of-service clauses (no "competing projects") to push customers out. > - **Contested and hedged.** The narrator is explicit and repeated: "There's no definitive proof." He frames the conclusion as inference — "it does seem like this is exactly what you would use the data for." He also concedes he "can't say all of this is actually happening." > - **Terms to define.** *Token maxing / spend maxing* — maximizing token consumption, often via agentic "looping." *Alpha* — a defensible competitive edge. *AI sovereignty* — Palantir's framing that an organization's data (and the insights compounding from it) is a strategic asset that should not be transferred. *Means of production* — the manifesto's Marxian flourish: your data is the input a competitor needs to reproduce your winning plays. > ### 🗒️ Notes > #### ⏱️ [00:00:00](https://www.youtube.com/watch?v=OQMPhkmeru0?t=0) Cold open — "Can you trust them?" > - Opens by asking whether "them" (the model providers) can be trusted. > - Introduces Palantir and Alex Karp with mockery, then pivots: the accusation of "shenanigans" coming *from Palantir* should make providers ask "are we the baddies?" > #### ⏱️ [00:02:00](https://www.youtube.com/watch?v=OQMPhkmeru0?t=120) Token maxing — the boat-store analogy > - Upsell framing: fishing boat → yacht → mega-yacht. > - Attributed Karp line: enterprises "chillax and waste time with tokens," get no value, give away IP. > - Aside mocking CNBC for "Twitch stream" / "Zoomer" language ("chillax," "shade"). > #### ⏱️ [00:05:00](https://www.youtube.com/watch?v=OQMPhkmeru0?t=300) Alpha and the manifesto's point three > - Two meanings of "alpha"; the relevant one is defensible edge / trade secrets. > - Quotes manifesto point 3 on "disposable scripts over robust software" and "false progress." > #### ⏱️ [00:08:00](https://www.youtube.com/watch?v=OQMPhkmeru0?t=480) "Safe" and the pricing question > - Karp clip on "safe" = not touching your learning data, not caching/replicating your busine Can you trust them? No, seriously, can you trust them? To understand what I'm talking about, you first must know who Palanteer is. Now, if you're not familiar with Palanteer, Palanteer, best known for surveilling um the whole world and Every Last Person, yeah, that company uh led, by the way, by Alex Karp, which can do this sweet move, and also is known for being geographically monogamous, which is in fact the definition of not being monogamous. Well, that Alex Karp of Palunteer has called the AI industry effing insane in a 20inut kind of uh crash out on CNBC. But here's the thing is that he made a lot of good points. Okay. And can you trust them? The them in this situation, the model providers, can you actually trust them? I don't know. I'm watching you, Dario. Also, can we sidebar here for a second? So, if if you're being accused of shenanigans by Palanteer, I feel like you got to, you know, take a moment and ask yourself some pretty tough questions. >> Are we the baddies? >> So, we're going to actually look at some of this kind of crash out, this mental breakdown that Alex Karp had on CNBC. And shockingly, I actually agree with a whole bunch of it. And even more importantly, the things he actually said, well, I actually think they they're happening. They're they have happened. Like, what he is saying is actually correct. And I'll show you exactly what I'm talking about after we talk about Alex Karp himself. So, first we got to say thank you to the sponsors. I've personally conducted hundreds of interviews and I know how hard hiring is. And that's not even with today's problems of fake AI profiles, resumes that are difficult to read, and senior engineers who have never even read code. G2I fixes that. Not the reading code part, the hiring part. G2I can help you hire frontend, backend, and even AI engineers because they have prevetted 8,000 plus engineers through real technical interviews so you can review quality candidates in days, not months. Check out g2i.co/prime and take the headache out of hiring. All right, so the first thing he talks about is token maxing. Now, if you're not familiar with token maxing, just imagine you go into a boat store and you're like, "Yo, I would like to get a fishing boat." And the fishing boat store was just like, "Yeah, but why why get a fishing boat when you can get a yacht, right? Like you can fish on a yacht." You know that, right? And you're like, "Well, I don't know about that." And they're like, "Yeah, but why even get a 100 foot yacht when you can get a 300 foot yacht?" Like, you get a mega yacht. You get like the world's biggest yacht. Like, why why would you ever want a fishing boat when you can get a yacht? To us, that'd be quite obvious what they're trying to do. They're trying to sell the most amount of money. But for some odd reason, when it comes to token spend, spend maxing as the kids call it, we just have zero care. This is I'm not throwing shade at them, but something has gone completely wrong. And the basic view among enterprises in this country is I'm gonna chill lax and waste my time with tokens. I'm going to get no value and they're going to get my IP. >> That sounds like shade. >> First off, it's very strange for me to watch the news. Yeah, this this little news segment right here. Why? Why is this sounding like a Twitch stream? Okay, who is saying chillax and shade on CNBC? I want the business news to be the business, not some sort of weird Zoomer millennial coded language. What's happening to the news? But this was just an unusual experience. But what he said was so dang correct. First off, token maxing and getting little value. It is obvious the value you get out of token maxing. When you incentivize employees to just use as much as you possibly can, that is not good. Because really, as he kind of alluded to, like the really big kind of piece of feedback wasn't the fact that you're spending a whole bunch and getting little value. It's the fact that what you're spending is on all of your business strategy. You are handing over to Claude. You're handing over to OpenAI all of the alpha, as they keep saying, by the way. The alpha uh has two meanings obviously. The first one is like your defensible position as a company, your trade secrets, the things that make your company be able to withstand or out compete your competitors. And the second version of alpha is a word you hear and then you know that whatever comes next is going to be the dumbest thing you've ever heard because you're talking to a crypto bro. Now in this case it's the it's the former not the latter. And of course with this crash out palanteer also released a 9point manifesto. Now I don't know if you know this but manifestos when it comes to tech pretty much always lead to good outcomes. I can't think of a single bad thing that has happened. But if you look at point three on the manifesto, the Palunteer manifesto, you'll see that token maxing hijacks your value orientation and decreases institutional fortitude and intelligence. The pursuit of high token usage incentivizes disposable scripts over robust software with the addictive feeling of false progress. It is absolutely an incredible and insightful point right there. I love it. And as somebody who is actively trying to do my own token maxing, my own looping to really kind of understand what these zoomers keep talking about, I just seem like I'm spending all my all my all my tokens and I'm not really getting much value out. But the argument for token maxing uh at least from his perspective, you'll notice that it wasn't just about the spend. It really was about handing over your IP. And he says that best here. >> So safe because it doesn't touch your unlearning data. Safe because it prevents the large language model from caching your data. and replicating your business. Uh uh safe because it doesn't transfer your IP of how to fight secret data, top secret data or in in a clinical context. >> He keeps using the term safe there. What he's talking about is the usage of these models. How do you make it safe? Well, the reality is to make it safe, the actual safety that we all need to be pursuing is that you need to know who's storing your data. Like as a company, are you giving away the secrets? are giving it all away to these these companies to just go off and and be able to use all of your hard-fought wisdom against you. Now, you're probably thinking, "Okay, that's a little ridiculous." Is it ridiculous? Is it? No company would ever do that. That's not actually happened. Well, just hold on. We got one more little part I want to go over and I think I can prove that that notion wrong. Okay, buddy? Cuz here's the deal. If what these big companies were selling was actually just able to derive value immediately, right? you didn't need the token max into potentially just dead end spending just a ridiculous amount of money. Seriously though, like look at Meta $73.7 trillion. They're claiming it's a two plus billion a year bill that they're spending on AI. Like what the what are they even doing over there? You can't tell me they're getting $2 billion worth of value out of their tokens right now. Nope. Refuse to believe it. Unless of course they're doing exactly what we're talking about, which is getting some information, getting that sweet data. >> If it was so valuable, let's say I can make you a billion dollars tomorrow. Wouldn't I say, "I'll make you a billion dollars and I want 30%." Why are they charging for tokens if it's so valuable? >> So good. Why are they charging for tokens if it's so valuable? Like that is a very big question because like just just really just walk with me on this. If AI was so freaking amazing that it's going to take whatever revenue you have now and triple it, you just have to integrate it into your uh into your system. Why would Anthropic even have a consumer side? They wouldn't. All they would do is simply be like, "Yo, business A, here's the deal. We're going to come in. We're going to triple your revenue. You're going to give us 30%." The end. Yeah. And you would say, "Absolutely. Yes, sir. Very much. Please. Thank you very much. I would absolutely love that." I would I would hands down do like no company would say no to that. I mean, other than Anthropic themselves, considering they're currently losing, they they would triple their losses. Okay, but besides for them and say OpenAI, like the rest of the companies would clearly jump on say absolutely, I would love to triple my revenue and I would gladly pay you for that. And that's of course it's because it's about the data, the secrets, the hard-fought knowledge, the alpha. I don't like that term. All right, so now is the part where I show you real world examples. you know that thing, which by the way, if I were to tell you that these companies were doing things that were shady, I think a lot of people would just instinctively go, "Yeah, I I could probably see that coming." Because the reality is, if you have all these users, millions upon millions of users making queries into your system, you're going to be able to see shapes of data in which most people will never be able to see. You can go, "Okay, this is really, really hot. This is what the people want. If we go into this industry, we're going to win." Well, look at this right here inside Cursor's Wild Rise. A lot of great new details. CEO Michael Truel didn't pay himself for years. Curser once made about 40 to 50% of Anthropic's revenue. Anthropic told Curser that Claude Code was just a research effort. Blah blah. Yeah. You know, you know old Claude Code, that old research effort. It's not like it's nothing. It's just like, uh, we're just trying something out. We're not like actually using all this data we've gathered. Where'd they get the data from? Hey, by the way, thanks for all the data. We really genuinely appreciate it. We're not like using it all. identifying the fact that there's millions upon billions of dollars sitting on the table here and we're going to just simply move into that industry. There's no definitive proof that Anthropic used their position, gathered all the data, and said, "Okay, this is the exact market we should go into." But it's a point on the graph. Okay. Oh, by the way, just in case you're wondering, our terms of service often includes things like, "Hey, you can't build competing projects or else we're going to drop you." And of course competing projects being Claude AI, Claude Pro and other products and services that we may offer for individuals along with any associated app software or websites together known as our services. Of course this is the consumer side of the toos. The commercial toos is worded slightly different. They have different affordances. But this David Sax tweet right here. I didn't know about another case. You don't think that can happen? That being, hey, you give away all of your information. Just look at Figma. Yes. If you haven't looked at Figma, by the way, their stock. Yeah, I'm so Hey, Figma employees, I feel for you. Okay, Anthropic employees, I don't feel for you. Look at that. Anthropic, your stock looks so good. But man, Figma absolutely decimated. Anyways, when you jump back here and say just look at Figma, according to the information, Anthropic blindsided its its then business partner with the launch of Claw Design. Figma's founder said Anthropic had not been consistently honest with them. Anthropic chief product officer. Anthropic chief product officer had even served on Figma's board until three days before the launch of Claw Design. I did two fingers up. Three days. I meant to do threes, not twos. Either way, just think about that for a second. If you're on the board for a company, you're there for the best interest of that company. We can all agree to that. Now, whether you're working there, if you're just an individual engineer, that's a completely different story. If you're on the board, your goal is to shepherd the company long-term toward success. The fact that if if all of this stuff that is being reported by the information is correct, which it is saying that Anthropic blindsided its business partners, the fact that Anthropic's chief product officer knew this was going to happen, knew Claw Design was going to drop and remained on the board for so long just seems completely crazy to me. I just can't even imagine that. That is like being not just stabbed in the back, you're being stabbed in the front. You're getting right just sitting there looking in your eyes and stabbing you and you don't even know about it until three days after they quit. Think about all of that juicy data they got from Figma 2. The data must have been absolutely fine. Like a nice delicious French wine. Okay, I'm not I don't Do French people drink wine? I assume so. This seem like they drink the wine. There's not even a good French accent. I don't even know how to do a fringe accent. I don't even know what that accent is. But now we have two points on a graph and two points on a graph makes well a line. And if you zoom out just a bit, you will notice that Anthropic also stole the world's books. So it's not like this is the first time that they have been inside of the data stuff. And plus, if you really zoom out and think about it, they also stole the world's knowledge. I mean, that's a that's a pretty big straight line if you ask me. Now, there's no definitive proof or like papers written up how they used Figma's data and then said, "Oh, okay. Yeah, hey, that's us now. We're we're now Figma." Instead, I think it's pretty obvious to say that they saw the success of Claude Code. They saw them moving into the coding space. They saw how popular the design space is that's paired with the coding space and said, "We need to move into that." And they could see the usage already. They already had the proof. They already had the pudding to know if they could just put the little toes again. Also, the pudding and the toes. That's that's not an analogy. It's just not one. They probably have many other things that are in the works that is being designed off the data that is being handed to them so that they can commandeer parts of the industry and then effectively be able to push people out with the terms of service. This is a very unusual experience. Now, I can't say all of this is actually happening or what is all their motivations behind things, but it does seem like this is exactly what you would use the data for. You would use it as a means to determine which industry you're going to capture. Because if you have unlimited token spend, you can effectively slop together any product you want to be able to be close enough with its competitors because you know that the most amount of people are in this realm. So, you can use all of your insider information. You can use all that hard fought knowledge. You can see how people are doing stuff in an industry and you can build the product that best matches other people's desires because you have the data. All the data. You have every single prompt and every single answer, every single reprompt. It's crazy what they actually have available. Now, this Alex Karp clip goes on for about 20 minutes. And yes, it's absolutely legendary some of the things he has to say. You should definitely watch the thing. I really am on Alex Karp's side. I can't believe I'm saying that. I'm on the uh geographically monogamous side of Alex Karp, which wait I shouldn't say it that way. That that's not that's not what it meant. I am definitely on the geographically monogamous Alex Karp's side of this argument, which is that the AI industry is effing insane because the effing insane part isn't the fact that they're losing a bunch of money, which I think is what everybody points to right now. It's the complete destruction of so many companies and the data they're able to yield and potentially able to weaponize against other people to be able to capture industries. Very, very interesting. And again, the nine-point manifest on AI sovereignty probably says it the best. Data retention is your treasure. Transfer it at your own peril. Your ability to win is dictated by your ability to recognize and use your unique edges. And you keep winning by compounding the underlying data to generate new insights. Transferring that data hands over access to your pre-existing winning plays and yields the means of production for new ones. Meaning somebody else can figure out how to win your industry based on your data. The name is the primogen., Planitar’s Manifesto, youtube.com, 01kwn1hy9cpyewn6j0bze90026.