Pakistan warns against autonomous military AI: the debate over weapons without human control gains political weight

By The Express Tribune · October 25, 2026.

According to a headline in The Express Tribune, Pakistan is warning against the use of autonomous artificial intelligence in the military sphere. The headline is all we have. Without the article itself, we can't say where the warning was made, who it was addressed to or what arguments the Pakistani government put forward. We'd rather not guess.

Broadly speaking, autonomous weapons systems can select and attack targets without direct human input. International forums have debated how to regulate them for years, but no binding treaty exists.

Our take: when a country like Pakistan, which faces a tense security situation, calls for caution, the debate has clearly moved beyond research labs and civil society groups. It is now a foreign policy issue. The position cuts both ways. A state may genuinely believe the risks of escalation and of mistakes nobody answers for are real. It may also be seeking diplomatic leverage over rivals with stronger technology. Without the full text, we can't say which explanation fits.

As we argue at Zendoric, the governance of military AI is one of the near-term risks that most needs attention. Hypothetical superintelligence is a long way off. The urgent issue is automating lethal decisions, and regulation there should rest on proven capabilities, not panic. Clear rules, such as meaningful human control over the use of force, would help protect the path toward AI's long-term benefits. We'll come back to this once verifiable details of the proposal emerge.