Emotion Recognition in Education Under the EU AI Act: Prohibited
Emotion recognition in education is prohibited under EU AI Act Article 5(1)(f), not high-risk. The ban applies since 2 Feb 2025; fines up to €35M or 7%.
This is one of the use cases people most often misfile. AI that infers the emotions of students in an educational setting is not "high-risk to be documented" — it is prohibited under Article 5(1)(f) of the EU AI Act, Regulation (EU) 2024/1689. There is no compliance route; the answer is not to deploy it. This page draws the line clearly, because emotion AI elsewhere can be lawful.
The prohibition (Article 5(1)(f))
Article 5(1)(f) prohibits AI systems used to infer emotions of a natural person in the areas of workplace and education institutions, except where used for medical or safety reasons (for example, detecting fatigue in a safety-critical context). So a tool that claims to read student engagement, attention, stress or mood from faces or voices in a classroom or exam is banned. The prohibition has applied since 2 February 2025, and breaches sit in the top penalty tier: up to €35 million or 7% of total worldwide annual turnover (Article 99(3)).
Why education is singled out
The Act treats workplaces and educational institutions as settings with a power imbalance, where people cannot freely opt out and emotion inference is especially intrusive and unreliable. That is why emotion recognition there is banned outright, rather than merely regulated as high-risk.
Where emotion AI is not prohibited
The ban is specific to workplace and education. Emotion recognition in other contexts is generally not prohibited, but may be high-risk as a biometric use under Annex III point 1 and/or carry an Article 50(3) transparency duty (telling people they are subject to emotion recognition). So "emotion AI" is not uniformly banned — it depends on where and how it is used. In education, it is banned.
The neighbouring high-risk education uses
Plenty of education AI is lawful but high-risk under Annex III point 3: systems used to determine admission or assignment to institutions, to evaluate learning outcomes, to assess the appropriate level of education, or to monitor and detect prohibited behaviour during tests. These carry the high-risk provider/deployer regime — they are not banned. Keep them separate from the emotion-inference prohibition.
Worked example: an ed-tech vendor
Scholaris builds an exam platform and considers adding a feature that flags "disengaged" or "anxious" students from webcam analysis during tests. That feature is prohibited under Article 5(1)(f) — it cannot be shipped in the EU, regardless of documentation. Scholaris instead keeps its automated proctoring for prohibited behaviour (Annex III point 3, high-risk) and its grading-support feature (also Annex III point 3), both lawful with the high-risk stack, and drops the emotion feature entirely.
How Confir helps
Confir tests an education use against the Article 5(1)(f) prohibition first, so an emotion-inference feature is flagged as prohibited rather than mis-scoped as high-risk, and it routes genuinely high-risk education uses (Annex III point 3) to the provider/deployer stack. The engine is deterministic and rule-based, citing the exact provision that fired.
Frequently asked questions
Is emotion recognition in education banned under the EU AI Act?
Yes. Article 5(1)(f) prohibits AI used to infer emotions of a person in the workplace or in education institutions, except for medical or safety reasons. The ban has applied since 2 February 2025, with fines up to €35 million or 7% of worldwide turnover. There is no compliance route — the use must be avoided.
Why is education treated more strictly than other settings?
Because educational institutions involve a power imbalance where people cannot freely opt out, and emotion inference is intrusive and unreliable. The Act bans it there outright rather than regulating it as high-risk, as it does some emotion AI in other contexts.
Is all emotion AI prohibited?
No. The ban is specific to workplace and education. Emotion recognition elsewhere is generally not prohibited but may be high-risk as a biometric use under Annex III point 1 and carries an Article 50(3) transparency duty. Context decides.
What education AI is high-risk rather than banned?
Under Annex III point 3: systems for admission or assignment to institutions, evaluating learning outcomes, assessing the appropriate level of education, and monitoring for prohibited behaviour during tests. These are lawful but high-risk, carrying the provider/deployer obligations.
What is the penalty for deploying it?
Emotion recognition in education is a prohibited practice in the top tier — up to €35 million or 7% of total worldwide annual turnover under Article 99(3). SMEs and start-ups are capped at the lower of the percentage or fixed amount under Article 99(6).
Related guides
- Article 5 prohibited practices
- Annex III high-risk use cases
- Predictive policing: prohibited or high-risk?
- How AI risk classification works
- EU AI Act for education
Manage your EU AI Act compliance in one place
Confir automates risk classification, technical documentation, and audit trails for any company. No consultants. No 6-month projects. 14-day free trial.
Start free trial →