The Dawn of the Age of AI Assassinations, By Brian Simpson
The dawn of the age of AI assassinations is upon us, and the events unfolding in early March 2026 mark a stark, irreversible milestone. The reported assassination of Iran's Supreme Leader Ayatollah Ali Khamenei — executed through a combination of years-long cyber infiltration, AI-driven pattern-of-life analysis, and precision strikes — represents not just another chapter in the long shadow war between Israel and Iran, but a fundamental shift in how state actors can project lethal power across borders.
Recent reports, including a detailed Financial Times investigation (widely cited across outlets like CNN, The Times of Israel, The Jerusalem Post, and others), describe how Israeli intelligence (primarily Mossad and Unit 8200) compromised Tehran's extensive network of traffic cameras over several years. These mundane urban surveillance tools became a goldmine: feeding vast streams of visual data into AI systems that mapped Khamenei's routines, his bodyguards' movements, travel routes, and the daily rhythms of his inner circle. By tracking proxies — like vehicle patterns near his compound on Pasteur Street — rather than the leader directly, operators built an intricate "patterns of life" profile without constant physical presence in Tehran.
This AI processing — analysing billions of data points from cameras, mobile networks, and other sources — allowed pinpoint accuracy. When a trigger event aligned (with alleged CIA confirmation of Khamenei's location), Operation Roaring Lion (or Epic Fury in U.S. terminology) launched: 30 precision-guided Sparrow missiles struck his compound in a daylight assault on February 28, 2026, killing Khamenei along with over 40 senior officials, including top military and intelligence figures. Cellular disruptions prevented warnings, and the strike drew on prior escalations, including Israel's 12-day war with Iran in June 2025 and the post-October 7, 2023 strategic decision to eliminate key architects of threats.
This wasn't a one-off; it builds on precedents like the 2020 AI-assisted remote sniper rifle killing of Iranian nuclear scientist Mohsen Fakhrizadeh (confirmed by The New York Times), where facial recognition and real-time compensation for movement delays enabled a satellite-linked machine gun to fire without a human operator in the loop. Israel's use of AI targeting tools (such as Habsora/Lavender in Gaza operations) has already drawn scrutiny for scaling decisions at machine speed, often with minimal human review. Now, applied at the highest level against a head of state, it demonstrates how AI lowers barriers: turning ubiquitous civilian infrastructure into intelligence assets, accelerating target acquisition, and enabling deniability or speed that human operatives alone could not achieve.
We are entering an era where assassinations become:
Scalable and remote — Algorithms sift petabytes of surveillance data faster than any analyst team.
Persistent and predictive — "Patterns of life" models forecast movements, reducing the need for risky on-ground assets.
Hybrid and multi-domain — Cyber hacks + AI analysis + kinetic strikes (missiles, drones) form a seamless kill chain.
Democratising lethality — If nation-states wield this, non-state actors or rogue elements will follow as tools proliferate (cheap drones, open-source AI, hacked IoT networks).
The implications are profound and chilling. For regimes like Iran's, paranoia will skyrocket — every camera, phone, or connected device becomes a potential betrayal. For democracies, the ethical line blurs: when AI nominates targets and humans merely approve (or not), accountability dilutes. Escalation risks multiply in a world where decapitation strikes can now be planned over years and executed in minutes. Iran's vows of vengeance, cyber retaliation (including hijacking cameras), and proxy activations across the region signal that this "dawn" may quickly become a prolonged, shadowy night.
This is no longer speculative fiction. The traffic cameras of Tehran, once symbols of everyday order, became instruments of regime change. The age of AI assassinations has begun — quietly, surgically, and with terrifying efficiency. The question now is not if it will spread, but how far, how fast, and who will be next in the crosshairs of the algorithm.
https://www.ft.com/content/bf998c69-ab46-4fa3-aae4-8f18f7387836
