AI sustainability is the effort to reduce the environmental and social costs of artificial intelligence. In practice, that means building and running models with less energy and water, less hardware waste, and fairer access to the systems they power. In cyber security, this matters because AI is increasingly used for threat detection, phishing analysis, fraud scoring, and incident response. A system that is effective but expensive to operate, hard to audit, or available only to large organizations can create new risks even when it improves security.
Attackers can also exploit unsustainable AI deployments. Heavy model use can be abused to drive up cloud bills, exhaust compute resources, or pressure shared infrastructure. Defenders respond with smaller models, rate limits, usage monitoring, and measurable efficiency targets so security tools remain resilient and affordable. Sustainable AI is therefore not just a climate concern; it is part of secure, responsible system design and governance.