Algorithmic decision support is software that helps a human make a choice by ranking, filtering, scoring, or recommending outcomes. It does not replace the decision-maker; it shapes what they see and what they are likely to approve. In cyber security, this matters because these systems often sit on a trust boundary: they may prioritize alerts in a SOC, score login risk, flag suspicious transactions, or sort vendor submissions for review.
Because the tool influences decisions, attackers may try to manipulate its inputs, bias its rankings, or exploit weak review processes. For example, a phishing campaign may be designed to evade a spam filter, or an adversary may flood a queue to bury high-risk events. Defenders need logging, validation, and human oversight so the recommendation path can be audited. Used well, algorithmic decision support improves speed and consistency; used badly, it can hide errors behind a veneer of automation.