Human-centric AI is artificial intelligence designed to support people, not replace their judgment or override their rights. It aims for transparency, accountability, privacy, and meaningful human control. In cybersecurity, this matters because AI systems increasingly mediate authentication, monitoring, content filtering, and decision support, so their design can affect user autonomy, data protection, and trust.
In practice, human-centric AI appears in defenses such as privacy by design, data minimization, explainable outputs, and human review for high-impact decisions. It also helps limit abuse when attackers use AI to profile users, manipulate behavior, or collect more data than a service needs. A human-centric approach reduces dependence on opaque platforms and makes it easier to audit access, retention, and model behavior. In short, it keeps security tooling aligned with fundamental rights instead of treating users as raw material.