AI unit economics measures how much business value an AI system creates for each dollar spent on model access, compute, storage, data retrieval, monitoring, and governance. In practice, it asks whether a workflow saves more time, reduces more risk, or increases more revenue than it costs to run. For security teams, this matters because AI features can look impressive but become expensive when scaled across users, requests, and controls.
In cyber security, unit economics appears in both defense and attack. Defenders use it to decide whether an AI triage assistant, phishing classifier, or log-analysis tool is worth its operating cost. Attackers also care about economics: cheap automation can flood systems with prompts, credential attempts, or malware variants until defenses are forced to spend more per event than the attacker does. Good security design improves unit economics by routing routine work to cheaper models, reducing manual review, and keeping high-cost AI focused on high-value cases.