Does my route-optimisation AI fall under the AI Act?
Adopted 2026-06-16 · ≈ 2 min read · 
edited by Dirk Baaijen
An AI that purely optimises routes or trips is usually not a high-risk system under the AI Act. If the same AI also drives task allocation or workers' performance, it does fall under Annex III. AI literacy already applies now.
Short answer: Usually not. An AI that only optimises routes, stop sequencing or deliveries to save cost, time or fuel does not appear on the high-risk list in Annex III. As soon as the same AI intervenes in workers' employment situation — who gets which trip, task allocation or performance management — the classification shifts to high-risk.
Why route optimisation usually falls outside high-risk
The AI Act (Regulation (EU) 2024/1689) takes a risk-based approach. An AI system is high-risk only if it is on the Annex III list or is a safety component in a regulated product (Annex I). A system that is purely operational — computing the most efficient sequence of stops, planning vehicle loads or predicting arrival times — does not touch any of the Annex III categories. It makes no decisions about individuals, creditworthiness, access to services or employment relationships. For such systems the heavy high-risk obligations (risk management system, technical documentation, conformity assessment) do not apply.
When does it shift to high-risk?
The line is drawn by the impact on people, not by the product name. Annex III, point 4, designates AI for employment-related decisions as high-risk: systems used for task allocation based on individual behaviour or personal characteristics, or for monitoring and managing work performance. A route optimiser that also automatically decides which driver runs which trip, or that scores and steers individual drivers' performance, touches this category directly. The actual functioning and the consequences for individual workers are decisive. A tool that computes routes and leaves the allocation to a human planner is more likely to stay outside the high-risk category.
What applies in any case?
Even if your system is not high-risk, some duties apply. The AI literacy obligation (Article 4) has been in force since 2 February 2025: anyone deploying AI must have a sufficient understanding of how the system works and what its limits are. Prohibited practices (Article 5) also already apply — for example emotion recognition in the workplace. For systems that are high-risk, the Annex III obligations were scheduled for 2 August 2026 but are being postponed to 2 December 2027 via the Digital Omnibus. A conservative approach is to document the actual functions of your system so you can substantiate its classification.
Read more: AI Act: timeline of obligations. Take the scan.
Sources
- https://eur-lex.europa.eu/eli/reg/2024/1689/oj
Regulation (EU) 2024/1689 (AI Act); risk classification in Article 6 and the high-risk list in Annex III. - https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai
European Commission policy page on the AI Act and its risk-based approach.
Read next
AI for demand forecasting and planning: does it fall under the AI Act?
AI for demand forecasting falls under the AI Act, but is usually not a high-risk system. If the same planning drives decisions about staff, Annex III may apply. Function and impact are decisive.
Is my planning algorithm high-risk under the AI Act?
Software that drives rosters, trip planning or recruitment falls under Annex III of the AI Act and counts as high-risk. The heaviest duties apply from 2 Dec 2027; AI literacy already applies now.
AI for dynamic freight pricing: what does the AI Act say?
AI that sets freight prices dynamically is not as such on the AI Act's high-risk list. AI literacy and transparency duties do apply, and the classification can shift where there are effects on individuals.