I want to point out a few obvious key points. (1) Destroying all the oil infrastructure in the Middle East is madness and could lead to an unprecedented global economic collapse and humanitarian crisis. We're talking about everything from food to gasoline; prices will rise, and inflation is already high. All of this will be blamed on the instigators of this war. (2) The pre-war situation before February 27, 2026, was actually quite good, but we are unlikely to return to that state. Everything in the Middle East is expected to suffer indefinite, long-term, and continuous destruction. (3) Furthermore, we must assume that technology financing will collapse for a very long time to come. The "genius" plans that plunged the Gulf states into war have already destroyed most of the funding for limited partners, data centers, and initial public offerings (IPOs). Any tech professionals who supported this war will likely soon experience the meaning of "force majeure" as their funds are withdrawn. (4) Many capital allocators will instead invest in lower levels of Maslow's hierarchy of needs, such as basic necessities like food and energy. (5) Fortunately, those progressives once proclaimed a “climate crisis.” Yes, their timelines were flawed, and much money was wasted on corruption, but the outcome was correct: we all need to wean ourselves off Middle Eastern energy as quickly as possible. Equally fortunate is that Elon Musk and China have taken climate change very seriously. Now, they need to immediately ship billions of solar panels, electric vehicles, batteries, nuclear power plants, and so on, to wean everyone off oil. (6) Of course, this isn't just about oil and gas. It's also about fertilizers and chemical precursors. Perhaps some new resources will become available at new prices, but scaling up takes time, especially at this scale, so shortages are almost inevitable. Nevertheless, China has actually scaled up coal-to-chemicals (C2C) [a,c] and there's an even more futuristic technology called “Power-to-X” [b], which can convert any energy, water, and air into hydrocarbons. But all of this needs to be accelerated. I have a background in chemical engineering, so I might start funding projects in this area. (7) Ultimately, this war will result in a heavy burden of guilt on everyone involved. Destroying such a massive infrastructure for the benefit of so many people is destined to be a lose-lose situation. Whatever their initial goals may have been, it will be a losing proposition. But unless you are truly capable of stopping this madness, the pragmatic approach is to do everything in your power to mitigate the damage this war will cause to yourself and your business.

Did you know that socialists don't understand capitalism? Well, nationalists don't understand diplomacy either. Socialists believe their country has absolute power domestically, while nationalists believe their country has absolute power globally, and therefore can seize anything they want by force because it's so powerful that nothing can happen and no one can resist. Both worldviews are flawed, for similar reasons. Socialists hate capitalists to the core, and nationalists hate foreigners to the core, often to the point that they are blinded by hatred. They begin to believe that as long as maximum force is used, foreign groups are no threat and are useless. Sometimes, this mentality takes them exceptionally far! The extent to which they plunder or conquer far exceeds your imagination. But ultimately, they will encounter constraints. By the way: Imperialist nationalists and self-sufficient nationalists do exist. Imperialists want to conquer other countries, while self-sufficient nationalists want nothing to do with them. Both actually have their logic and fallacies. The logic of imperialism is that, like any enterprise, it will decline if it doesn't develop. But it also has a fallacy: the fallacy of excessive imperial expansion. Similarly, the logic of self-sufficiency is that self-reliance is a good thing. Its fallacy lies in the fact that when self-sufficiency goes to extremes, it negates the inherent division of labor in capitalism, or the idea of learning from others. Generally speaking, it's best to think of these political paradigms as programming paradigms. Just as sometimes a recursive approach is needed instead of an imperative approach, sometimes a self-sufficient approach is needed. But you don't need to define yourself as a recursor or a self-sufficient individual. Just use the best tools. Theoretically, there is a third path, neither expansionism nor self-sufficiency. That is to be a "normal" nation, trading with other nations, neither conquering nor hating them! However, in practice, this is not the case. This means you are either part of an empire or you are independent enough to survive outside of it.

If Iran wins, five eras will come to an end. 1991-2026: The Unipolar Era 1974-2026: The Petrodollar Era 1945-2026: The Post-War Era 1776-2026: The Alliance Era 1492-2026: The Western Era Specifically, the end of the petrodollar era (1974) will also mark the end of the unipolar era (1991) and the post-war order (1945). It will mark the moment when Eurasian powers once again dominate Western powers (1492). Finally, the rapid decline in the purchasing power of the dollar, coupled with military defeats, could very well lead to the collapse of the American alliance (1776). Few seem to truly understand the extent of America's dependence on currency issuance. But the end of the petrodollar era also signifies the end of Keynesianism as we know it. If the cost of living suddenly skyrockets and political polarization approaches civil war levels, we might see scenarios described by Dalio, *The Fourth Turning Point*, and Turchin. Below are some charts illustrating the increasing polarization in American politics. Whether you look at congressional voting or individual statements on social media, polarization is on the rise. 2015: PLOS ONE 2016: Vox 2017: CJR 2017: Pew Research Center 2020: Turchin

Software development is no longer just a technical job; it's a creative practice. Introducing Replit Agent 4, the first AI specifically designed for creative collaboration between humans and intelligent agents. Design on an infinite canvas, collaborate with teams, run multiple agents in parallel, and deliver working applications, websites, slideshows, and more.

All those who preach principles need to understand that this is all about tribalism. Democrats support the Starlink project to aid Biden's military operations but oppose Antropico serving Trump's military operations. Antropeak supported data duplication when training models but opposed it when the models were subjected to Chinese data distillation attacks. Republicans oppose state power when Democrats use it to promote the "Awakening" movement, but support it when Republicans use it to promote the "Make America Great Again" (MAGA) movement. I mean, need I say more? It all makes perfect sense. It's as logical as supporting your own gun and opposing someone pointing a gun at your face. That said, recognizing the realities of tribalism doesn't mean opposing cooperation, or even principles. You can cooperate within a tribe. You can even cooperate between different tribes. You just need to understand which tribe you belong to and which tribe you're dealing with. You can also stand by your principles. But understand this: only principles that bring collective strength to the group over time will survive natural selection. Once you understand your own group, and the other group, understand your principles, and understand the other group's principles… you can propose a truly win-win solution that satisfies the interests of both sides. Just don't expect the other group to be foolish enough to agree to unilaterally abandon principles proposed by your group (selectively!). Because if following the rules leads to failure, what's the point of the rules? Until recently, Americans have begun to adopt this tribal mindset. Why? Because Democrats assume everyone agrees with their universalist ideology; they pride themselves on being "untribalized." Meanwhile, Republicans, accustomed to easily controlling global hegemony, are not accustomed to tolerating any genuine resistance. In the tech world, there's a huge divide between artificial intelligence and cryptocurrency. The cryptocurrency mindset naturally carries a tribal flavor. This is the essence of cryptocurrency tribalism. But AI technologists have become incredibly wealthy due to their political naiveté. Remember: they were always burying themselves in their work, without ambition. In the San Francisco Bay Area, even after events like the COVID-19 pandemic and the Black Lives Matter movement, they persisted in starting businesses because they believed the post-war center-left consensus would return. They believed the winds would shift, everything would pass, and so on. It was this admirable yet naive belief—the belief that America would remain stable—that drove them to build everything they created there, seeing it as Silicon Valley's final, glorious contribution to the world before its complete demise. Today, political conflict, wealth taxes, and attacks on data centers and tech founders are escalating. Artificial intelligence is learning a lesson that cryptocurrencies have long understood—digital is politics, and the dialectical relationship between the network and the state is inescapable. Therefore: suppose Silicon Valley as a geographical region is declining because it simply cannot simultaneously contend with the Democratic Party, the Republican Party, and China. Suppose this decline would radiate to thousands of startup cities around the world. Then, turn your attention to the realm of decentralization.

Red, Gray, Blue The biggest mistake most commentators make today is talking about the "United States of America" as if it were still a unified whole. In reality, it's more like fragmented tribes in North America. Different American tribes now have their own preferred opinion leaders, foreign policies, gender identities, corporations, counties, and even currencies. The only thing they don't fully possess is their own nation. But massive migrations between blue and red states have already occurred, and the digital divide has begun, with people flocking to different social networks. It's clear that the endless online conflicts will eventually spill over into reality; what awaits us is a formal division of America. Until then, we remain trapped in this strange gray area, where people still talk about "American policy." However, we all know that when discussing "South Korea policy," it's crucial to immediately clarify whether you're referring to North Korea or South Korea. Similarly, when discussing "American policy," you must first clarify whether you're referring to Blue America, Red America, Tech America, or their increasingly numerous sub-clades. Silicon Valley vs. the Pentagon This leads to the ongoing conflict between Silicon Valley AI companies and the Pentagon. This is just one of many conflicts between the internet and the state, and one of many lose-lose situations unfolding on the chessboard between blue states, red states, and the gray/tech camp. But you can't understand what's happening without looking at it from a camp perspective. In short: Center-left tech people at Anthropologie (and many at OpenAI and Google) say they don't want their software used for autonomous weapons and domestic surveillance. They see themselves as protecting civil liberties. Meanwhile, center-right tech people in the U.S. government want to use AI to defend the nation. They resent the idea that a single tech company can arbitrarily veto any military plan (that might involve sensitive information). They see themselves as protecting national security. Tribal Perspective This debate about principles is untenable because it's not really about principles, but about fundamental control. Anthropologie employees trust their CEO and believe he can handle user data properly. They don't trust the "Make America Great Again" (MAGA) supporters at the Pentagon. The Pentagon's intuition is the opposite: they don't want these wavering "awakened" individuals controlling their military and president. Fundamentally, the tech left doesn't want the tech right (not to mention the Republicans) to have fundamental control over them, to take over their companies, or to monitor them. The tech right, in turn, doesn't want the tech left (not to mention the Democrats) to have fundamental control over them, to veto their military plans, or to obstruct their presidency. All of this occurs against the backdrop of numerous other intense conflicts, including the conflict between the Democrats and the tech community (through wealth taxes) and the conflict between the tech community and the Democrats (through artificial intelligence disrupting blue-collar jobs). The tech right thinks the tech left is incredibly foolish for not seeing that the tech left is their only shield from heavy Democratic taxes (not to mention being exploited by China). The tech left, in turn, thinks the tech right is incredibly foolish for wanting to create a surveillance state that could ultimately fall into Democratic hands… and become like China. This leads to the China issue. The China Challenge Both camps constantly view China as an outlier. Clearly, China is developing its own weapons. Therefore, unless the US government takes the same measures as China, its military strength cannot surpass China's! On the other hand, China also conducts domestic surveillance. Therefore, if the US government were to adopt the same measures as China, its military strength would also be unable to surpass China's! The two camps also have differences on the issue of China. The left wing of tech companies, including Anthropico, uses the framework of the US-China confrontation to secure AI funding. They are also angry that China stole their models. Therefore, there is indeed some anti-China sentiment within the tech left. Conversely, the tech right, including many government officials, uses the argument that China monitors its citizens to oppose censorship. They are also angry that the Democratic Party abuses state power to suppress them. Therefore, there is indeed some internal liberal tendency within the tech right. (Note: The far left and far right oppose technology and all overseas military operations for different reasons, and therefore they are not directly involved in this debate.) Defeat However, ultimately, all of this is irrelevant. At the current rate, the US government simply cannot win the competition against the Chinese government because the American people are unable to cooperate for the greater good; therefore, a true "United States of America" does not exist. Modern Americans champion freedom (red), protest (blue), or techno-capitalism (gray). While these are all admirable, they emphasize individual rights rather than the collective responsibility that Americans held in the 1950s. Meanwhile, the Chinese pursue harmony, the Party, and techno-communism. They have constructed a social contract that merges the state, the regime, and the internet into a massive "Voltron." For China: their nationalists are Republicans, their statists are Democrats, and their tech elite are Silicon Valley. Of course, internal conflicts exist, but these are currently suppressed by the Party. The result is that the Chinese have collectively built what is perhaps the most powerful manufacturing giant on Earth. This Chinese-style fusion of nation, regime, and internet challenges both the Republican Party's faith in freedom and the Democratic Party's belief in democracy. Technologists are deeply committed to founder-led capitalism. Simply appealing to principles cannot counter China's "Voltron." It's as futile as praying to Zeus to ward off a nuclear bomb. You need a set of principles that truly generate collective power, a force capable of matching China's. Otherwise, you can only surrender. Surrender or cooperation? Democrats (and the broader Western left) were actually the first to realize this.This is why figures like Carney, Mamdani, Newsom, and Walz ultimately surrendered to the Chinese government. The Biden-led Democrats pulled out all the stops against China, but the Chinese government ultimately prevailed and proved itself the superior force. Therefore, Democrats are now (effectively) selecting overseas bureaucrats for Xi Jinping's empire, advisors to communist Canada and Chinese California. The tech community and Republicans haven't fully grasped this. Republicans still claim their military is superior to China's, and Silicon Valley still believes its technology is superior to China's. They point to a few areas where they still hold an advantage, attempting to ignore China's immense advantage in scale and speed (especially in the physical world). You might think that mass-produced, somersault-capable "astronaut" kung fu robots would be enough to wake both the red and tech camps. But demonstrations may not be enough; we may need to witness the Chinese drone fleet in action to truly understand the market. Perhaps it will debut in Ukraine, Iran, or another proxy war. As Orwell said, "False beliefs will eventually collide with solid reality."

After four years of almost uninterrupted emergence of new AI tools, it's clear this trend isn't slowing down anytime soon. Therefore, we need some kind of ranking algorithm. @amasad has a dynamic feed on Replit containing thousands of constantly updated mini-apps ranked by usage. @shayne_coplan could assemble a panel of experts and have Polymarket provide odds for the panel's votes (e.g., "Which model will these 100 tech experts choose as the best image model in May 2026?"). @nikitabier could integrate this functionality into X, similar to Consumer Reports, using ranking signals from an algorithm. Google could do this too, but they'd likely rank all their own products first (while Gemini performs exceptionally well). There are many possible ideas. But it feels like software development is moving at least 3x faster, maybe even 10x faster. Therefore: fast, reliable directories and rankings become even more important. In today's world of rapidly growing web pages, they might be as valuable as early Yahoo and Google were.

Artificial intelligence is fantastic for custom software targeting small markets. Indeed, its impact is most significant the smaller the market, because small markets typically cannot afford the costs of software development.

AI Tribes Can I offer a perspective on AI that is neither all-or-nothing? Argument: AI can boost productivity within trusted tribes, but it can diminish economic trust between tribes. Therefore: Within a tribe, you can share information, search archives, and operate more digitally and independently than ever before. In time, you might even be able to write scripts to have bots build various real-world items for you. However, between tribes, we will face a deluge of AI spam and scams. From AI sales and marketing calls to AI spam websites and fake AI resumes, the list goes on. (While there are some cryptographic countermeasures to address this, it ultimately comes down to the trust issue of Web3. That is, you must trust the person who uses cryptography to assure you of the authenticity of an AI photo.) In short, AI may significantly boost productivity within a tribe, but it could also significantly diminish trust between tribes. The costs of verification, oversight, and censorship are skyrocketing. Therefore, artificial intelligence may disrupt almost as many markets as it creates. China may be the primary exception to this rule because it can centrally manage the Chinese internet, thus preventing AI-driven fraud. Therefore, it has the potential to possess the world's largest community of cooperatives in the digital age. Democrats can't do this because they hate AI; Republicans can't either because they hate Chinese-style digital boundaries. Furthermore, there's fierce competition between Democrats and Republicans. Therefore, in the age of AI, it's unlikely that the US will have a large, trustworthy "American" community. To borrow a popular internet phrase, the US isn't even a well-functioning economic zone. However, following the same logic, AI does indeed make small cooperative communities far more productive than ever before. A highly collaborative startup community could be far more productive than a larger, more decentralized traditional nation. So: the future may see a thousand Estonias and Chinas, a thousand networked nations with millions of people, and a superpower with a billion people. Not zero, not infinity, but one thousand and one.

Artificial intelligence is on the inside, encryption is on the outside. Within the trust boundaries of your community, you only need to provide the AI with all the information it needs, and it can make its own judgment. But outside these trust boundaries, you will encounter a large amount of spam from AI. At this point, you need encrypted verification to distinguish between true and false information.