What Is Automated Decision-Making?

Imagine applying for a loan online and receiving a rejection within seconds. No bank employee reviews your application, asks additional questions or explains the decision. Instead, an algorithm evaluates your information and determines the outcome. Automated decision-making can make processes faster and more efficient, but what happens when these decisions affect people's rights and opportunities?
What Is Automated Decision-Making?

Inhaltsverzeichnis

What Is Automated Decision-Making?

Automated decision-making refers to decisions made by technological systems with little or no human involvement.

Such systems may rely on traditional software, statistical models or artificial intelligence. They are used in areas ranging from financial services to recruitment and insurance.

A related concept is profiling, which involves automatically processing personal data to evaluate or predict characteristics such as someone’s economic situation, behaviour or personal preferences.

However, profiling and automated decision-making are not identical. Profiling does not necessarily result in a decision, and automated decisions do not always involve profiling.

When Does the GDPR Restrict Automated Decisions?

Article 22(1) of the General Data Protection Regulation (GDPR) establishes a general prohibition on decisions based solely on automated processing where they produce legal effects or similarly significantly affect individuals.

An automatic rejection of a loan application or a job application may fall within this protection.

Three conditions are particularly important: the decision must concern an individual, be based solely on automated processing and have sufficiently significant consequences.

The involvement of a human does not automatically remove a decision from Article 22. Human intervention must be meaningful rather than a formality.

Nor does the provision apply only to sophisticated AI. Conventional algorithms can also fall within its scope.

When Are Automated Decisions Allowed?

Article 22(2) provides three exceptions.

Solely automated decisions may be permitted where they are necessary for entering into or performing a contract, authorised by EU or Member State law providing suitable safeguards, or based on the individual’s explicit consent.

These exceptions are narrowly defined. Simply obtaining consent or including automated processing in contractual terms does not automatically make a decision lawful.

For the contractual necessity and explicit consent exceptions, individuals must receive safeguards including the right to obtain human intervention, express their views and contest the decision.

Additional restrictions apply when special categories of personal data are involved. Organisations must also comply with the GDPR’s other requirements, including having an appropriate legal basis for processing.

What Have European Courts Decided?

In its 2023 SCHUFA judgement (C-634/21), the Court of Justice of the European Union clarified that automated credit scoring can fall within Article 22 where a third party relies heavily on the score when making a contractual decision.

In Dun & Bradstreet Austria (C-203/22), decided in 2025, the Court strengthened the right to receive meaningful information about the logic underlying automated decisions.

Individuals must be able to understand and challenge how their personal data was used to reach a particular result.

Final Thoughts

Automated decision-making raises a question that extends beyond technical efficiency: how much control should individuals retain over decisions affecting their lives?

European data protection law does not prohibit automation altogether. Instead, it establishes limits where automated decisions carry significant consequences.

The challenge is ensuring that human oversight provides genuine accountability rather than merely the appearance of control.

Sources & References

Stay curious and let’s keep exploring the future of law together!

This article was written with the assistance of various AI tools.

Related Article

50 AI Agents Tried to Build a Civilisation in Minecraft. They Invented Bureaucracy Instead.

50 AI Agents Tried to Build a Civilisation in Minecraft. They Invented Bureaucracy Instead.

Legal Design: Making Law Work for People

Legal Design: Making Law Work for People

What Is Legal Engineering? Bridging the Gap Between Law and Technology

What Is Legal Engineering? Bridging the Gap Between Law and Technology