War, AI, and State Power: Why the Iran Crisis and the Anthropic Clash Signal a New Era of Global Strategy

Author: Zion Zhao Real Estate | 88844623 | 狮家社小赵 | wa.me/6588844623

Author’s note and disclaimer: For general education and market literacy only. Not financial, investment, legal, accounting, or tax advice, and not an offer, solicitation, or recommendation. Information is general and may be inaccurate or change. No liability accepted. Investing involves risk, including loss of principal; past performance is not indicative of future results. This article is written based on one of my favorite podcast, The All-In Podcast. 

The New Architecture of Power: Iran, Artificial Intelligence, and the Battle for Strategic Control

The conversation featuring Emil Michael, the current U.S. Under Secretary of War for Research and Engineering, is valuable not because every claim in it can be accepted at face value, but because it exposes three decisive transformations in contemporary statecraft. First, war is now inseparable from data, drones, and software-defined command architectures. Second, frontier artificial intelligence has become a governance problem before it has become a settled technology problem. Third, military advantage increasingly depends not only on battlefield prowess, but also on cloud contracts, insurance backstops, critical minerals, batteries, and procurement design. The podcast therefore deserves to be read not merely as commentary on one war or one corporate dispute, but as a window into a deeper reordering of power among the state, the technology sector, and the industrial base (U.S. Department of Defense, n.d.; Reuters, 2026i; Reuters, 2026j). (U.S. Department of War)

The immediate geopolitical setting is real and grave. Reuters has reported that the United States and Israel launched strikes on Iran, that Ayatollah Ali Khamenei was killed in the opening phase of the campaign, that the Pentagon’s operation has been publicly identified as Operation Epic Fury, that six U.S. reservists were killed in a drone attack on a U.S. facility in Kuwait, and that a U.S. submarine sank the Iranian warship Dena off Sri Lanka. Those are not speculative talking points. They are the hard, verified frame around which the podcast’s wider arguments unfold (Reuters, 2026a; Reuters, 2026c; Reuters, 2026k). (Reuters)

Yet the podcast is much less reliable when it moves from facts to certainty about war aims and political end states. Public reporting after Khamenei’s killing did not support a neat conclusion that regime collapse would rapidly follow. Instead, Reuters reported a high-stakes succession struggle, the possibility of clerical and Revolutionary Guard consolidation, and strong institutional incentives inside the Iranian system to preserve continuity even under extraordinary stress. In other words, the removal of a supreme leader is not the same thing as the removal of the governing order. That distinction matters analytically, because modern coercive campaigns often succeed tactically while failing politically when they mistake elite decapitation for regime transformation (Reuters, 2026b). (Reuters)

That same need for precision applies to the nuclear question. The podcast speaks in a manner that implies an imminent, settled, and fully visible Iranian nuclear threat. Publicly available evidence is more complex. In its February 2026 safeguards report, the International Atomic Energy Agency stated that Iran had previously accumulated an estimated 440.9 kilograms of uranium hexafluoride enriched up to 60 percent U-235, but also made clear that, after hostilities and the withdrawal of inspectors for safety reasons, the Agency could no longer verify the current size, composition, or whereabouts of Iran’s stockpile, nor confirm whether enrichment-related activities had fully ceased. The prudent conclusion, therefore, is not false certainty, but heightened proliferation concern under conditions of degraded visibility (International Atomic Energy Agency [IAEA], 2026).

One of the podcast’s most ambitious and controversial claims is that the Iran campaign should be read partly through the lens of China. This is more plausible than many critics admit, but less conclusive than the speakers suggest. China did set a lower 2026 growth target of 4.5 percent to 5 percent, signaling a weaker macroeconomic posture than in prior eras of Chinese expansion. China also bought more than 80 percent of Iran’s shipped crude in 2025, representing about 13.4 percent of its seaborne crude imports, while Venezuelan oil accounted for roughly 4 percent of China’s seaborne imports. Those facts do mean that pressure on Iran and Venezuela can create indirect leverage over Chinese energy security. But the stronger leap in the podcast, namely that such pressures straightforwardly imply Chinese invasion incentives or a near-inevitable Taiwan scenario, remains a strategic inference rather than an empirically demonstrated conclusion (Reuters, 2026e; Reuters, 2026f; Reuters, 2026g). (Reuters)

The more compelling part of the discussion concerns the changing grammar of war itself. The podcast’s emphasis on drones is consistent with a growing scholarly literature showing that the most consequential military uses of artificial intelligence today are often not fully autonomous “Skynet” scenarios, but intelligence fusion, targeting assistance, pattern recognition, sensor integration, and decision acceleration. Anthony King’s analysis of “digital targeting” is especially relevant here: he argues that the dominant literature has overfocused on spectacular visions of autonomous weapons, while the more immediate transformation lies in AI-enabled military intelligence and target development. That distinction helps explain why the podcast’s examples oscillate between drone swarms, missile defense, object discrimination, and edge inference rather than autonomous sovereign decision-making in the pure sense (King, 2024). (DOI)

At the ethical level, the podcast also gestures toward a real scholarly divide. David Deptula argues that autonomous weapon systems, if responsibly designed and employed by operators working within the laws of armed conflict, can enhance rather than degrade ethical force application. Mary Ellen O’Connell argues the opposite, namely that autonomous weapons raise such serious legal and moral concerns that prohibition is justified. The significance of the podcast is that it shows this debate migrating from journals and policy forums into procurement terms, cloud architecture, and real contract disputes. What used to sound abstract is now becoming operational and commercial (Deptula, 2023; O’Connell, 2023). (DOI)

The Pentagon–Anthropic clash is therefore not a sideshow. It is one of the most important state–technology confrontations yet to emerge in the frontier AI era. Reuters reported that the Defense Department told Anthropic it was a “supply-chain risk,” while broader federal procurement rules were being tightened to require “all lawful uses,” irrevocable rights, and an absence of intentionally embedded partisan or ideological judgments in government AI systems. Anthropic, for its part, publicly maintained that it would not abandon red lines around mass domestic surveillance and fully autonomous weapons. Meanwhile, OpenAI’s subsequent defense arrangement preserved explicit restrictions on mass domestic surveillance, autonomous weapons targeting, and high-stakes automated decisions. The result is not a simple morality tale of one side being pro-safety and the other side being anti-safety. It is a constitutional struggle over who sets binding limits when private models become public infrastructure (Reuters, 2026h; Reuters, 2026i; Reuters, 2026j). (Reuters)

Anthropic’s own public materials make the dispute more intelligible. The company states that Claude’s constitution is a detailed expression of the values and behavior Anthropic wants the model to embody, and that this constitution directly shapes Claude’s behavior. That is a remarkable fact. It means the model is not merely a neutral technical artifact awaiting downstream policy. It is already normatively structured at the model layer. For ordinary commercial use, that may be a feature. For sovereign military use, it becomes an argument about democratic legitimacy, delegated authority, and mission assurance. The state’s concern, in that frame, is not just censorship or inconvenience. It is dependency on a privately governed reasoning substrate whose behavioral boundaries are partly authored outside the chain of public command (Anthropic, 2026). (Anthropic)

This is where the best governance scholarship becomes indispensable. Hickok argues that public-sector AI procurement carries a special burden because public institutions do not merely optimize for efficiency, but also for rights protection, accountability, due process, and legitimacy. The OECD similarly warns that AI procurement requires governance structures robust enough to avoid opacity, overreliance, and lock-in. Alhosban, Pesingu, and Kalyanam show that vendor lock-in is not merely a commercial nuisance, but a multidimensional dependency problem involving technical, contractual, and economic frictions. Read together, these sources illuminate the real lesson of the Anthropic episode: the central risk was not simply that one company had strong views, but that the government had allowed a frontier capability to become deeply embedded before governance, fallback options, and model pluralism were fully secured (Alhosban et al., 2024; Hickok, 2024; OECD, 2025). (MDPI)

The podcast is also strongest when it links AI to procurement reform. Howell, Rathje, Van Reenen, and Wong show that more open defense R&D competitions generate more commercial innovation and more military adoption than conventional, tightly specified procurement approaches, which tend to create lock-in without equivalent technological gains. That finding mirrors Emil Michael’s remarks about moving away from requirement systems that overdetermine solutions before firms can experiment. It also aligns with long-standing lessons from DARPA. Bonvillian argues that the DARPA model remains a uniquely powerful innovation architecture, while Carlsson describes DARPA as one of the most successful experimentally organized innovation systems in the world. The broader implication is that defense modernization is not just about spending more. It is about designing institutions that can absorb uncertainty, reward iteration, and scale what works without being captured by legacy process (Bonvillian, 2018; Carlsson, 2025; Howell et al., 2025). (IDEAS/RePEc)

The industrial dimension of the podcast deserves equal attention. The Strait of Hormuz carried roughly 20 million barrels per day of petroleum liquids in 2024, or about one-fifth of global petroleum liquids consumption, according to the U.S. Energy Information Administration. Reuters reported that war-risk premiums surged and that the Trump administration ordered support for tanker insurance and up to about $20 billion in Gulf maritime reinsurance to stabilize flows. This matters because finance is not external to war. Insurance is an instrument of logistics, and logistics is an instrument of strategy. In this sense, the maritime insurance discussion in the podcast is not a digression. It is a reminder that modern coercion operates through shipping lanes, underwriting capacity, risk pricing, and sovereign balance-sheet credibility as much as through aircraft carriers and strike packages (Reuters, 2026d; U.S. Energy Information Administration, 2025). (Reuters)

The same is true upstream in the supply chain. Official U.S. defense reporting on China describes a historic military buildup across naval, space, missile, cyber, and nuclear domains, with Beijing seeking a “world-class” military by 2049. The Pentagon’s China report emphasizes the scale of the People’s Liberation Army’s modernization and the breadth of China’s shipbuilding and strategic weapons base. Against that backdrop, Michael’s emphasis on batteries, critical minerals, and manufacturing capacity is analytically sound. The defense-industrial contest is now inseparable from questions of mineral processing, energy storage, electronics, and production finance. States do not dominate twenty-first-century competition merely by inventing faster. They must also extract, refine, manufacture, deploy, and replenish faster (Department of Defense, 2024).

What, then, should be the final judgment on the podcast’s substantive thesis? Its strongest insight is that war, AI governance, and industrial policy have collapsed into a single strategic field. Its weakest tendency is to convert that insight into overconfidence about motives, timelines, and political end states. The Iran war may indeed reshape bargaining power with China, but that does not make every battlefield outcome legible as grand strategy. The Pentagon may indeed be right to reject operational dependence on a model provider whose terms it views as unstable, but that does not eliminate legitimate democratic concerns about surveillance and autonomy. And frontier AI firms may sincerely claim ethical restraint, yet that does not settle the question of whether privately authored constitutions should govern public missions of national defense. The deeper truth is harder and more consequential: the emerging world order will be shaped by whoever learns to govern these intersections without surrendering either strategic effectiveness or public legitimacy. (Reuters)

In that sense, the podcast is historically important even where it is imperfect analytically. It captures a transitional moment when military power is no longer reducible to troop counts, ship tonnage, or even munitions stockpiles alone. Power now resides in model access, cloud control planes, procurement flexibility, drone production, mineral security, insurance capacity, and the ability to align public authority with private technical systems under crisis. That is the true subject of the discussion. Iran is the immediate theater, Anthropic is the immediate controversy, but the larger issue is the institutional architecture of state power in the age of algorithmic war. (U.S. Department of War)

References

Alhosban, A., Pesingu, S., & Kalyanam, K. (2024). CVL: A cloud vendor lock-in prediction framework. Mathematics, 12(3), 387. https://doi.org/10.3390/math12030387 (MDPI)

Anthropic. (2026, January 22). Claude’s new constitution. https://www.anthropic.com/news/claude-new-constitution(Anthropic)

Bonvillian, W. B. (2018). DARPA and its ARPA-E and IARPA clones: A unique innovation organization model. Industrial and Corporate Change, 27(5), 897–914. https://doi.org/10.1093/icc/dty026 (DOI)

Carlsson, B. (2025). Experimentally organized innovation systems: Lessons from DARPA, Apple, Amazon, Google, and Operation Warp Speed. Industrial and Corporate Change, dtaf054. https://doi.org/10.1093/icc/dtaf054 (OUP Academic)

Department of Defense. (2024). Military and security developments involving the People’s Republic of China 2024. U.S. Department of Defense. (U.S. Department of War)

Deptula, D. A. (2023). An operational perspective on the ethics of the use of autonomous weapons. Ethics & International Affairs, 37(3), 261–271. https://doi.org/10.1017/S0892679423000266 (DOI)

Hickok, M. (2024). Public procurement of artificial intelligence systems: New risks and future proofing. AI & Society, 39, 1213–1227. https://doi.org/10.1007/s00146-022-01572-2 (DOI)

Howell, S. T., Rathje, J., Van Reenen, J., & Wong, J. (2025). Opening up military innovation: Causal effects of reforms to US defense research. Journal of Political Economy, 133(11), 3605–3651. https://doi.org/10.1086/737235 (IDEAS/RePEc)

International Atomic Energy Agency. (2026, February 27). NPT safeguards agreement with the Islamic Republic of Iran(GOV/2026/8).

King, A. (2024). Digital targeting: Artificial intelligence, data, and military intelligence. Journal of Global Security Studies, 9(2), ogae009. https://doi.org/10.1093/jogss/ogae009 (DOI)

O’Connell, M. E. (2023). Banning autonomous weapons: A legal and ethical mandate. Ethics & International Affairs, 37(3), 287–298. https://doi.org/10.1017/S0892679423000357 (DOI)

OECD. (2025). AI in public procurement: Governing with artificial intelligence. Organisation for Economic Co-operation and Development. (OECD)

Reuters. (2026a, March 1). Iran’s supreme leader Khamenei killed, Iranian state media confirm. Reuters. (Reuters)

Reuters. (2026b, March 1). Khamenei killing shatters Iran’s order, triggers high-stakes succession race. Reuters. (Reuters)

Reuters. (2026c, March 5). Pentagon identifies two soldiers killed in Iran war. Reuters. (Reuters)

Reuters. (2026d, March 3). Trump orders oil tanker insurance support, says Navy could escort ships in Gulf. Reuters. (Reuters)

Reuters. (2026e, March 5). China sets 2026 economic growth target at 4.5%-5%. Reuters. (Reuters)

Reuters. (2026f, January 13). China’s heavy reliance on Iranian oil imports. Reuters. (Reuters)

Reuters. (2026g, January 7). Chinese refiners expected to replace Venezuelan oil with Iranian crude, traders say. Reuters. (Reuters)

Reuters. (2026h, February 28). OpenAI details layered protections in US defense department pact. Reuters. (Reuters)

Reuters. (2026i, March 6). US defense department told Anthropic it is a supply chain risk, CEO says. Reuters. (Reuters)

Reuters. (2026j, March 7). US draws up strict AI guidelines amid Anthropic clash, FT reports. Reuters. (Reuters)

Reuters. (2026k, March 4). US sub sinks Iranian warship off Sri Lanka, killing dozens. Reuters. (Reuters)

U.S. Department of Defense. (n.d.). Emil Michael. https://www.defense.gov/About/Biographies/Biography/Article/4232659/emil-michael/ (U.S. Department of War)

U.S. Energy Information Administration. (2025, September 4). What is the Strait of Hormuz, and why is it so important to global energy markets?

From Battlefield to Cloud Stack: What Iran and Anthropic Reveal About Modern Power and Sovereignty

Macroeconomics matters to Singapore property clients because global conflict, energy shocks, artificial intelligence disruption, and supply chain stress do not stay overseas. They flow directly into inflation, interest rate expectations, business confidence, hiring trends, rental demand, and investor sentiment. For buyers, that affects affordability, loan planning, and the timing of entry. For sellers, it shapes pricing strategy, buyer psychology, and how to position an asset in a cautious market. For landlords and tenants, it influences rental resilience, expatriate demand, corporate relocation patterns, and lease negotiations. For investors, it reinforces a core truth: real estate is not just about location, but also about macroeconomics, capital flows, and risk management.

In an increasingly complex world, Singapore property decisions can no longer be made by looking at real estate in isolation. Capital flows, interest rates, geopolitical risk, energy prices, currency movements, equity market cycles, technology disruption, and policy shifts all shape demand, liquidity, pricing power, rental resilience, and long term asset performance. That is why discerning clients should work with an advisor who understands not only property, but also the wider forces that move wealth.

In Singapore, where property decisions are closely tied to financing costs, policy stability, and global wealth movements, understanding these wider forces can help clients avoid emotional decisions and act with greater clarity. Periods of geopolitical tension often separate average decisions from strategic ones. The right move may not be to rush, delay, or follow headlines. It is to assess your position objectively, understand the market cycle, and align your property choices with your financial goals, time horizon, and risk tolerance.

As a Singapore real estate professional with deep grounding in macroeconomics, global affairs, asset allocation, portfolio construction, market analysis, and Singapore legal frameworks, I bring a broader and more strategic lens to every client conversation. My work is not transactional. It is analytical, disciplined, and rooted in due diligence. I dedicate substantial time every day to studying the economy, monitoring markets, writing essays, and distilling complex global developments into practical insights for property buyers, sellers, landlords, tenants, families, and investors. That daily commitment allows me to advise with context, clarity, and conviction.

For international families, China Chinese clients, Southeast Asian investors, ultra high net worth individuals, family offices, and institutions looking at Singapore for wealth preservation, relocation, education planning, family office set up, or long term capital deployment, this matters even more. Whether you are purchasing a primary residence, acquiring an investment property, securing a home for children studying in Singapore, planning for 陪读家庭 arrangements, or positioning capital in a stable and globally respected market, the right advice should extend beyond square footage and recent comparables. It should reflect financing conditions, policy risk, tenant demand, exit strategy, portfolio fit, and your broader wealth objectives.

Real estate deserves a serious place in a well constructed portfolio. Compared with many other asset classes, quality property can offer lower volatility, stronger asset backing, potential for long term capital appreciation, and recurring rental income similar in nature to dividend cash flow, while also serving practical lifestyle, education, and legacy planning goals. The key is not simply to buy property. It is to buy the right property, at the right time, for the right reason, within the right strategy.

If you value thoughtful advice, rigorous research, discretion, and a globally informed perspective on Singapore property, I would be honoured to assist you. Engage my services for a strategic, data driven, and client centric approach to buying, selling, renting, or investing in Singapore real estate. In a world full of noise, let us make your next property decision with discipline, insight, and confidence.

That is where I add value. I help clients translate complex global developments into practical Singapore property decisions, whether you are buying your first home, upgrading, rightsizing, selling for maximum value, securing quality tenants, or building a long term investment portfolio.

If you want clear, data driven, and strategy focused guidance for buying, selling, renting, or investing in Singapore property, engage my services today. Let us turn uncertainty into informed action and position you for the next opportunity.

Let us turn global complexity into Singapore property clarity.