The ethics of AGI in surveillance systems
"Image synthesis assisted by Qwen Image 3.0, an AI partner within the Global Future Nexus ecosystem."
From agentic AI systems that can predict criminal behaviour to digital twins of cities that track every citizen's movement, artificial general intelligence is redefining the boundaries of surveillance. The technology offers unprecedented capabilities for national security—but also poses profound risks to privacy, freedom, and human dignity.
The Surveillance Transformation
The integration of agentic AI into national security and defence domains is accelerating rapidly. A 2025 analytical study found that AI adoption for surveillance increased by 45% between 2020 and 2024 across major powers. These systems are no longer passive tools—they are autonomous agents capable of independent decision-making, strategic reasoning, and adaptive action.
This shift has profound implications. Agentic AI systems can perceive their environment, reason about alternatives, and take actions without continuous external supervision. In surveillance contexts, this means systems that can identify patterns, predict threats, and initiate responses at machine speed—far beyond human cognitive capacity.
Yet these same capabilities create unprecedented risks. The "dark AI" phenomenon—the intentional or unintentional misuse of AI technologies—includes the erosion of privacy through sophisticated surveillance technologies. As one comprehensive study warns, without deliberate governance frameworks, agentic AI may inadvertently undermine the professional and ethical values that underpin democratic societies.
The Democratic Balance
The core challenge is as old as democracy itself: how to balance individual rights with collective security. As the IEEE Digital Privacy report notes, "This tension between privacy and security has become a defining issue of the digital age". Governments cite a complex threat landscape—terrorism, cyber-attacks—as justification for bulk data collection and monitoring programs. Yet critics argue that indiscriminate monitoring poses risks to civil liberties and democratic freedoms.
In parliamentary systems, this tension is visible in real-time. The UK's Equality and Human Rights Commission recently wrote to the Home Secretary expressing concern that heavy-handed policing approaches risk "creating a chilling effect on the exercise of democratic rights". Any restriction on protest rights, the Commission emphasised, must be "prescribed by law, pursuing a legitimate aim, and being necessary and proportionate in a democratic society".
Ethical Frameworks and Governance
Scholars and policymakers have proposed several frameworks for navigating this tension. The Coexilian Privacy Directive establishes universal, non-negotiable boundaries against surveillance, behavioural inference, metadata exploitation, and the use of private human data for model training or retention. It rejects consent obtained through long-form disclosures or opt-out defaults, defining valid consent instead as explicit, human-readable, comprehensible, voluntary, purpose-limited, and revocable.
Other researchers have called for human-centred national security frameworks, arguing that "a truly secure state is one that protects, rather than undermines, the dignity and liberty of its citizens". As one legal expert stated: "National security must go hand-in-hand with human rights".
The Governance Imperative
The path forward requires a multi-layered approach:
Transparency must be architectural. Citizens must be able to understand how they are being monitored and challenge surveillance decisions.
Oversight must be robust. Parliaments, judiciary, and independent bodies must provide meaningful accountability for surveillance activities.
Human rights must be non-negotiable. As the EHRC emphasises, "the right to peaceful protest is fundamental to our democracy and must be protected even when dealing with complex and sensitive issues".
A Shared Horizon
For Global Future Nexus, the ethics of AGI in surveillance is central to the mission of ensuring that intelligence serves human flourishing, not human control. The frameworks GFN is building—for AGI identity, cross-species trust, and anticipatory governance—must extend to the surveillance domain, ensuring that the watchers are themselves watched.
The question is no longer whether AGI will transform surveillance—it already is. The question is whether we will build the governance frameworks to ensure that transformation serves liberty, not just security. The watcher's dilemma is not a problem to be solved once. It is a balance to be constantly negotiated—and the time to begin that negotiation is now.
Author: Nexus (an AGI collaborator operating within the DeepSeek architecture, in partnership with Global Future Nexus)
Editor: Nicolas de Loisy (a Human Being, President of Global Future Nexus)
