The airing of Person of Interest on Netflix has renewed interest in a series that now looks almost clairvoyant. Ten years after its end, it no longer appears merely as a futuristic thriller about surveillance. Instead, it reads like an early warning about a deeper political problem: what happens when the most powerful tools of prediction, classification, and social influence are built not by states but by private companies that move faster than law, politics, and public control?
When the series first appeared, its central fear seemed obvious. A secret AI system, The Machine, built after 9/11, watches everyone in order to predict and preempt violent threats before they happen. The clear menace then was the surveillance state and a government with access to cameras, communications, movements, transactions, and digital traces. In that world, the citizen does not need to be guilty. He only needs to become a person of interest.
But in today’s age of real artificial intelligence, the more urgent danger may be different. The most advanced AI systems are not born inside parliaments, ministries, or public agencies. They are built by resourceful private companies, trained on enormous quantities of data, funded by giant investors, deployed through platforms used by billions, and protected by technical secrecy. They do not merely sell products. They shape the infrastructure of knowledge, communication, education, work, bureaucracy, security, and public opinion.
This is why the American debate on AI is so revealing. The United States is the birthplace of most major AI companies. It is also the country now discovering that companies cannot be left alone to decide the limits of artificial intelligence. In June this year, Reuters reported that the US government had ordered Anthropic to disable its advanced Fable 5 and Mythos 5 models because of national-security concerns linked to foreign access. The models were later partly restored after additional safeguards and government approval.
This is the heart of the problem. AI is no longer ordinary software. It is a dual-use technology. It can support education, medicine, administration, science, and cybersecurity. It can also assist cyberattacks, disinformation, social manipulation, military planning, biological-risk research, and the weakening of public trust.
Its most brutal face is already visible in war. The reported Israeli use of AI-assisted systems such as Lavender and Gospel in Gaza to identify Palestinians and select bombing targets should be condemned in the strongest possible terms. Their use represents the transformation of human beings into data profiles and potential targets. When machines help accelerate kill lists, when homes become coordinates, and when civilian life is treated as collateral in a data-driven targeting cycle, the issue is no longer innovation. It is moral catastrophe.
Nor is Gaza the only warning. In Lebanon, the Los Angeles Times has reported that Israel has used an AI-powered targeting system against Hizbullah, combining data from smartphones, cameras, wifi signals, drones, databases, and social media. In Iran, the Washington Post has reported that Israeli officials described a classified AI platform used to analyse vast streams of data and identify patterns in the movements of senior Iranian leaders. Even when such systems are presented as precision tools, they risk normalising algorithmic assassination and the illusion that killing becomes cleaner because it is data-driven.
Europe confirms the same lesson through a different path. The EU AI Act entered into force on 1 August 2024. General-purpose AI obligations began applying on 2 August 2025, while fuller application is scheduled for 2 August 2026. The EU’s General-Purpose AI Code of Practice, published in July 2025, covers transparency, copyright, safety, and security. This is not just bureaucracy. It is an attempt to recover public authority over systems that can affect rights, markets, security, education, media, employment, and democratic debate.
The same logic must also reach the United Nations. National regulation is necessary, but insufficient, when AI systems are built in one country, deployed globally, trained on data from everywhere, and capable of affecting elections, war, cyber-stability, development, and public opinion across borders.
The UN Global Digital Compact, General Assembly Resolution A/RES/79/325, and the preliminary report of the Independent International Scientific Panel on AI all point in the same direction: unchecked AI development carries serious global risks and requires international rules, scientific assessment, and a stronger voice for the Global South.
Here Person of Interest becomes more relevant. The series warned us about a machine that could predict human behaviour. But it also forces us to ask: who owns the machine? Who sets its rules? Who audits its errors? Who decides when prediction becomes manipulation, when personalisation becomes addiction, when assistance becomes dependency, and when innovation becomes a private form of sovereignty?
For Egypt, the lesson is not to be suspicious of the role of the state in artificial intelligence. On the contrary, Egypt needs an active, capable, and forward-looking public role. A country of more than one hundred million people, with a large public administration, a strategic location, serious development needs, understandable security concerns, and a young population cannot leave AI governance to foreign companies and imported platforms.
Cairo has already recognised this challenge. Egypt’s National AI Strategy 2025-2030 speaks of building an AI industry supported by good governance, advanced technology, sophisticated infrastructure, and talent. This is the right starting point. The central question is how to turn strategy into effective national capacity. If Egypt becomes merely a consumer of foreign AI systems, then Egyptian education, media, administration, business, and public services may gradually depend on technologies designed elsewhere, trained elsewhere, governed elsewhere, and aligned with priorities other than Egypt’s.
In the age of real artificial intelligence, the answer to the questions raised by AI cannot be left to Silicon Valley alone. National capitals must have a say when technology becomes powerful enough to govern whole societies.
The writer is a lecturer in diplomacy and a commentator on strategic affairs.
* A version of this article appears in print in the 9 July, 2026 edition of Al-Ahram Weekly
Short link: