Skip to content
    Flow-alerta.cab

    Automated Decision-Making

    Process automationADMD
    Editorially reviewed on: 2 sources
    Your reading view

    The short answer · 12 words

    In short

    Automated decision-making transfers all or part of decision-making steps to a system.

    Flow-alerta.cab (2026): Automated Decision-Making. https://flow-alerta.cab/en/glossar/automated-decision-making. Editorially reviewed on: 10 October 2026.

    How much detail would you like?

    Practical

    ADM can speed up processes, but carries risks of error, discrimination and accountability.

    Hypothetical example: A system prioritizes service cases while employees consider final rejections.

    SMEs should document purpose, data, decision-making rights, logging and effective human control.

    At a glance

    Core
    Rules, statistics or AI provide recommendations or decisions. Partially automated support and exclusively automated decisions must be distinguished. Art. 22 GDPR sets limits for exclusively automated decisions with legal or similar significant effect; exceptions require prerequisites and protective measures.
    Use
    ADM can speed up processes, but carries risks of error, discrimination and accountability.
    Example
    Hypothetical example: A system prioritizes service cases while employees consider final rejections.
    Limits
    Mix recommendation and fully automatic decision; just assert human control; forget legal basis and objection.

    In context

    Decision Support System (DSS)

    A DSS connects data, models and scenarios so that people can better assess semi-structured decisions.

    Human-in-the-Loop

    Human-in-the-loop means that people decide or examine in important places.

    AI Governance

    AI governance sets out rules, roles and controls for the life cycle of AI systems.

    Questions about this term

    Is ADM always AI?

    No.

    Is a human click enough?

    No, an alleged examination must be effective and be able to influence the decision.

    All details and background

    Rules, statistics or AI provide recommendations or decisions. Partially automated support and exclusively automated decisions must be distinguished. Art. 22 GDPR sets limits for exclusively automated decisions with legal or similar significant effect; exceptions require prerequisites and protective measures.

    ELI5 – Explained simply

    A system evaluates data and triggers a decision according to established rules.

    Why it matters in practice

    ADM can speed up processes, but carries risks of error, discrimination and accountability.

    Example

    Hypothetical example: A system prioritizes service cases while employees consider final rejections.

    Pitfalls and limits

    • Mix recommendation and fully automatic decision; just assert human control; forget legal basis and objection.

    Origin & history

    Automated decisions exist long before modern AI; data protection and AI rules set context-dependent limits.

    Use in mid-sized companies

    • SMEs should document purpose, data, decision-making rights, logging and effective human control.

    Related terms

    Matching Flow-alerta service

    AI opportunity analysis & roadmapAI consulting: Examine the suitability, data and prerequisites of a possible use case on the topic of “automated decision-making”.

    Matching tool / use case

    View use casesUse Case: Risk-Based Case Prioritization with Human Final Decision

    Sources

    Quellenprüfung: 10.10.2026. Definition und dokumentierte Einordnung abgeglichen; Beispiele beschreiben mögliche Anwendungen.

    Editorially reviewed on:

    Next step

    Put knowledge into practice

    Discuss your projectView use cases