Fast autonomous military systems can interact in ways that produce escalation no human intended, even when each system stays within its own rules—creating a “flash-crash” risk for warfare.

Financial markets after 2010 retained automation but added system-level brakes (circuit breakers, price bands, audit trails); military AI needs analogous friction that increases with strategic consequence.

Practical safeguards include matching decision speed to escalation risk, contracting authority when evidence degrades, ensuring real human intervention windows, and creating rapid incident channels for algorithmic near-misses.

Financial markets learned to interrupt machine-speed cascades without abandoning automation. Military AI needs the same system-level brakes before a tactical interaction becomes a strategic fact.

The Pentagon is entering the final weeks of a 90-day rewrite of its policy on autonomy in weapon systems. President Donald Trump’s June 5 national-security AI memorandum ordered faster adoption across the defense and intelligence enterprise while directing the Department of Defense to update Directive 3000.09 so rapidly evolving AI systems remain consistent with the chain of command and operational authorities.

That rewrite should address a problem that sits between the individual weapon and the political decision: what happens when several fast systems, each behaving inside its own rules, interact quickly enough to create an escalation none of their human owners chose?

Financial markets encountered the same structural problem sixteen years ago. On May 6, 2010, the Dow Jones Industrial Average fell more than 9 percent from its previous close before rebounding with extraordinary speed. The joint SEC-CFTC investigation found that a large automated sale of E-mini S&P 500 futures entered an already fragile market, high-frequency traders rapidly bought and resold contracts, liquidity evaporated, and stress propagated across linked venues. No single machine had to become rogue for the system to produce a collective failure.

The........

© Eurasia Review