Fines for Improper AI Use in Companies: What the AI Act Brings
If you think data protection fines are high, wait until you see those from the EU AI Regulation. We are talking about penalties up to €35 million or 7% of global revenue, whichever is greater.
What type of AI can be fined?
The AI Act classifies systems by risk level. High-risk systems include those used in employment: recruitment, employee evaluation, dismissals and task assignment. Using algorithms to decide hiring or promotions carries strict obligations.
Prohibited practices
Social scoring of employees, emotion recognition at work and subliminal manipulation are directly prohibited.
What should your company do?
Inventory all AI-using systems. Classify them by risk level. Ensure high-risk systems meet transparency and oversight requirements. Document everything.
Time tracking, vacation management and document storage do not require high-risk AI. RegulaKit manages compliance without opaque algorithms or regulatory risks.
Frequently Asked Questions
Are there fines for misuse of AI in companies?
Yes. The EU AI Act allows penalties up to €35 million or 7% of global turnover for prohibited uses.
Which AI uses are prohibited?
Social scoring, manipulation, mass biometric surveillance, facial-image scraping and emotion recognition at work.
When does the AI Act apply?
It entered into force in August 2024 with a phased rollout. Prohibitions apply from February 2025; the rest by August 2026.
Disclaimer: This article is for informational purposes only and does not constitute legal advice. The information may not be complete, accurate or up to date. For specific legal matters related to your company, always consult a qualified labor law professional or your employment advisor.