Weekly Dispatch on AI, Geopolitics and Markets #15

Israeli and U.S. forces launched coordinated strikes on Iran last Saturday, hitting military bases, nuclear facilities, and leadership targets in Iran. Iran’s Supreme Leader Ayatollah Ali Khamenei was killed in one of the strikes on a fortified leadership compound in Tehran, triggering an emergency succession process inside the Islamic Republic’s leadership. Fighting has since spread across the region. Israeli and U.S. aircraft have continued striking targets across Iran, while Tehran and allied militias launched missiles and drones toward Israel and at U.S. bases in Gulf states, including Kuwait, Qatar, and the United Arab Emirates. Israel has also struck Hezbollah targets in southern Lebanon after the group entered the conflict, raising fears of a wider regional war.

It was a bold move. Washington likely expected a rapid decapitation effect — a leadership shock that could trigger political fragmentation inside Iran. But Iran is not Venezuela. The Islamic Republic has spent four decades building a deeply entrenched system with parallel military structures, ideological institutions, and a security apparatus embedded across the state and economy. That architecture was designed to survive exactly this kind of shock.

In the short term, the strikes will likely degrade Iran’s missile and drone capabilities. But the political objective set by the Trump administration — regime change — remains an extremely high bar. The Iranian system has historically absorbed severe pressure, from war to sanctions to internal unrest, while maintaining core regime continuity. Several outcomes appear likely in the coming weeks:

The most probable scenario is escalation without invasion. Air strikes, cyber operations, and missile exchanges could intensify, but a U.S. ground invasion remains unlikely. The strategy appears focused on degrading Iran’s military capabilities rather than occupying the country. At the same time, proxy warfare could expand. Iranian-aligned groups in Iraq, Lebanon, and Yemen have the capacity to open multiple smaller fronts through drones and missiles. Instead of one battlefield, the conflict could fragment into several, creating constant pressure and strategic fatigue across the region.

Energy markets are another pressure point. Even partial disruptions in the Strait of Hormuz or Red Sea shipping lanes could sustain higher oil and gas prices. Markets respond not only to closures but to uncertainty, meaning even intermittent disruptions can keep insurance costs, freight rates, and risk premiums elevated.

Inside Iran, internal turbulence is more likely than regime collapse. The leadership transition will trigger maneuvering among elites, but the security apparatus — particularly the Revolutionary Guard — remains intact, organized, and capable of enforcing stability. The more probable outcome is consolidation rather than collapse.

Eventually, the conflict may reach a forced diplomatic pause, as rising casualties and energy disruptions push regional powers and global actors to seek de-escalation. The key question now is how far each side is willing to push before the cost curve rises sharply. That will determine whether this evolves into a prolonged regional conflict or a short but violent shock.

But there is another risk. Any outcome short of regime change — particularly a negotiated accommodation with the regime’s new leadership — could alienate segments of the Iranian diaspora that expected a decisive break with the Islamic Republic. In that sense, even a strategic stalemate could create new political liabilities for Washington.

Last week, Block Inc. delivered one of the clearest signals that artificial intelligence is beginning to reshape corporate hiring. On Feb. 28, CEO Jack Dorsey said the fintech company would cut roughly 40 percent of its workforce, reducing headcount from more than 10,000 employees to fewer than 6,000. Dorsey argued that advances in intelligence tools now allow much smaller teams to build and run products. Investors welcomed the move, and shares jumped more than 20 percent in premarket trading.

The shift is now reaching Wall Street. On March 4, Morgan Stanley began cutting about 3 percent of its global workforce, or roughly 2,500 employees, according to people familiar with the matter. The reductions span investment banking, trading, wealth management, and investment management, despite the bank reporting record revenue in 2025. The timing suggests the move reflects structural efficiency pressures rather than cyclical weakness.

Last year, layoffs across the tech sector were rarely attributed to artificial intelligence. Most companies described them as post-pandemic corrections, overhiring during the boom years, or routine efficiency drives. Block’s announcement changes the tone. It is one of the first large companies to state the logic openly. AI allows firms to operate with significantly smaller teams. Once one company proves the model works the pressure on others to follow quickly grows.

Some analysts caution that AI’s impact on employment remains uncertain and argue that companies may be using the technology as a convenient explanation for broader cost-cutting. Bloated payrolls and slower growth still explain part of the downsizing in many industries.

The larger risk lies in timing. AI-driven productivity gains are arriving just as the global economy faces rising geopolitical volatility and macroeconomic uncertainty. Together, these forces create a perfect storm. Companies are under pressure to cut costs, and technology now allows them to do so faster, and a fragile global environment amplifies the consequences.

What might have been a gradual adjustment in the labor market could instead accelerate into a sharper employment shock. The shift may not be beginning. It may already be accelerating.

AI company Anthropic has entered a dispute with the U.S. Department of Defense over the military use of its models. The company declined to remove internal safeguards that limit the use of its AI in offensive military applications, including certain surveillance and autonomous weapons uses. The disagreement has complicated Pentagon efforts to integrate frontier AI systems into defense programs.

Following the standoff, the Department of Defense formally designated Anthropic a “supply chain risk”, arguing that national security systems require reliable and unrestricted access to critical AI infrastructure. The move could restrict the company’s technology from certain government and contractor deployments and highlights a growing conflict between frontier AI developers and governments over who ultimately controls how advanced models are used in military contexts.

Anthropic was initially selected because it offered one of the strongest technical solutions available. Limiting access to such models for political or procurement reasons introduces a potential national security risk. At the same time, Anthropic’s stance reflects a deeper tension shaping the AI industry. Allowing frontier models to be integrated into military systems without robust guardrails, particularly in autonomous or semi-autonomous weapons, poses serious risks to both developers and users. Once deployed in such environments, the margin for error becomes extremely small.

What looks like a short-term commercial setback may ultimately strengthen the company’s position. In an increasingly politicized AI landscape, drawing a clear line on safety and deployment standards can elevate a firm as a voice of restraint and responsibility. For policymakers, researchers, and enterprise customers, credibility carries real weight. This episode may become a defining moment in the debate over AI and national security.

The lifespan of companies in the S&P 500 is shrinking. Firms are entering and leaving the index more quickly. Several forces are accelerating the churn.

First, creative destruction is speeding up. Markets change faster than before, making it harder for companies to stay dominant for decades.

Second, technology keeps resetting the playing field. The internet, cloud computing, mobile platforms, and now AI allow new companies to scale at extraordinary speed — often overtaking incumbents before they can adapt.

Third, mergers, acquisitions, and private equity are removing companies from public markets more frequently, through buyouts or consolidation.

The result is clear: corporate longevity is falling. Humans may be living longer, but companies in the S&P 500 are surviving for shorter and shorter periods.

👦📖 Why Boys Are Falling Behind in Education
A thoughtful piece examining the widening gender gap in education and literacy. The article argues that boys are increasingly disengaged from traditional schooling models and calls for structural changes in how reading and learning are taught.
https://ofboysandmen.substack.com/p/boys-can-read

🌍 What Khamenei’s Assassination Means for Iran
Foreign Policy analyzes the strategic implications of Iran’s leadership crisis and how it could reshape the regional balance of power following the strike that killed the country’s supreme leader.
https://foreignpolicy.com/2026/02/28/iran-khamenei-ayatollah-assassination-israel-us-war/

🏛️ Anthropic vs. the Pentagon
Anthropic’s CEO discusses tensions between AI developers and the U.S. government over military use of frontier models — a preview of the policy battles likely to define the next phase of the AI race.
https://www.youtube.com/watch?v=MPTNHrq_4LU

⚙️ The Machine That Builds Machines
Chamath Palihapitiya outlines how AI-driven automation could transform manufacturing by enabling systems that design and produce the next generation of machines themselves.
https://www.linkedin.com/pulse/making-machine-builds-machines-chamath-palihapitiya-qirqc

📚 What AI Executives Tell Their Own Kids About the Jobs of the Future
A fascinating look at how leaders building the AI revolution are advising their own children about careers.
https://www.wsj.com/lifestyle/careers/what-ai-executives-tell-their-own-kids-about-the-jobs-of-the-future-1ba43f65