Artificial intelligence tools are reshaping how companies attract, assess, and manage talent. From CV screening software and automated interview scoring to performance monitoring dashboards and predictive analytics, AI is now embedded in the HR workflows of organizations worldwide.
For international employers operating in Germany or planning to do so, however, deploying these tools without legal preparation carries significant risk. Germany’s employment law framework – layered with European data protection rules, co-determination rights, and anti-discrimination obligations – places requirements on employers that are not found in many other jurisdictions.
This article outlines what international employers need to understand before introducing AI-driven tools into their HR operations in Germany.
AI Systems and the EU AI Act
The EU AI Act, which entered into force in 2024 and is being phased in progressively, establishes a risk-based framework for the use of artificial intelligence across the European Union. AI systems used in employment contexts – including those used for recruitment, candidate screening, performance evaluation, task allocation, and monitoring – are classified as high-risk systems under Annex III of the Act. This classification carries substantial compliance obligations.
Employers deploying high-risk AI systems must ensure that the systems are subject to adequate human oversight and cannot operate as the sole decision-maker on matters that affect individuals. Detailed technical documentation and records of how the system functions must be maintained. Transparency obligations require that individuals affected by AI-assisted decisions be informed that AI has been used. Before a high-risk system is deployed, it must undergo a conformity assessment to verify that it meets the Act’s requirements.
For international employers accustomed to lighter-touch regulatory environments, the compliance infrastructure required under the EU AI Act represents a meaningful operational commitment. Legal advice should be sought on current applicability, as obligations are entering into force in stages through 2026 and beyond. Legal advice should be sought on current applicability. The high-risk classification obligations under Article 6, which govern the employment and HR use cases covered in this article, enter into force on 2 August 2026.
The following table illustrates how the EU AI Act’s risk classification applies to common HR use cases, and the primary legal concerns each raises.
| AI Use Case | Risk Classification | Primary Legal Concern | Action Required |
|---|---|---|---|
| CV screening and candidate ranking | High risk | EU AI Act Annex III; Article 22 GDPR | Document selection criteria; establish human oversight; audit for bias |
| Automated aptitude or personality testing | High risk | EU AI Act Annex III; AGG | Ensure transparency and explainability; define appeal process; test for discriminatory outcomes |
| AI-assisted performance monitoring | High risk | Section 87(1) No. 6 BetrVG; Section 26 BDSG | Works council agreement required before deployment; document data processing basis |
| HR chatbot handling applicant queries | Limited risk | Article 50 EU AI Act | Disclose AI use to applicants; ensure handover to human contact is available |
| AI tools for drafting job advertisements | Minimal risk | AGG | Review output for discriminatory language before publication |
GDPR and Employee Data Protection in Germany
The General Data Protection Regulation (GDPR) applies directly and uniformly across the EU, but Germany adds a further layer through the Federal Data Protection Act (Bundesdatenschutzgesetz – BDSG). Together, these frameworks impose strict requirements on how personal data is processed in the context of recruitment and employment.
Under Section 26 BDSG, personal data belonging to employees and job applicants may only be processed where this is necessary for decisions relating to the establishment, performance, or termination of an employment relationship. The provision covers applicants and former employees as well as current staff. Where processing is based on consent, the inherent power imbalance in an employment relationship means that consent must be assessed carefully. Genuine voluntariness cannot simply be assumed, and the BDSG requires that consent be given in writing or electronically, with the employer informing the individual in writing of the processing purpose and their right to withdraw.
Article 9 GDPR is directly relevant where AI tools in recruitment process special categories of personal data, including data revealing racial or ethnic origin, health data, or biometric data. The processing of such data is prohibited as a default position, with narrow exceptions applying. Employers must ensure that AI-driven screening or assessment tools do not inadvertently collect or process such data without a lawful basis.
Of particular importance for AI-driven HR processes is Article 22 GDPR, which grants individuals the right not to be subject to a decision based solely on automated processing where that decision produces legal effects or similarly significantly affects them. Rejection of a job application following automated CV screening is precisely the kind of decision this provision is designed to address. Where automated processing is used, employers must implement suitable safeguards, including at minimum the right for the individual to obtain human intervention, to express their point of view, and to contest the decision.
Works Council Co-Determination Rights
Perhaps the most significant and least anticipated requirement for international employers is Germany’s system of workplace co-determination. Where a works council (Betriebsrat) exists – and one may be established in any private-sector company with at least five eligible employees – the employer cannot unilaterally introduce technical systems that monitor employee behavior or performance.
Section 87(1) No. 6 of the Works Constitution Act (Betriebsverfassungsgesetz – BetrVG) grants the works council a mandatory right of co-determination over the introduction and use of technical devices designed to monitor the behavior or performance of employees. This provision applies directly to the majority of AI tools used in HR contexts: performance management software, productivity tracking systems, AI-assisted appraisal tools, and similar technologies all fall within its scope.
This is not a consultation right – it is a co-determination right. The employer must reach agreement with the works council before the system is introduced. Where agreement cannot be reached, the matter is referred to a conciliation board (Einigungsstelle), whose decision is binding on both parties. An employer that introduces such a system without the works council’s agreement acts unlawfully and may be required to discontinue use of the tool.
International employers setting up operations in Germany, or acquiring existing German businesses, should establish at an early stage whether a works council exists and, if so, ensure that any planned deployment of AI-driven HR tools is subject to the required process. Where a works council does not yet exist, it is worth noting that employees retain the right to establish one at any time.
Equal Treatment and Anti-Discrimination Obligations
The General Equal Treatment Act (Allgemeines Gleichbehandlungsgesetz – AGG) prohibits discrimination in recruitment and employment on grounds including race, ethnic origin, sex, religion or belief, disability, age, and sexual orientation. The AGG applies to the entire recruitment process, including any stages at which AI tools are used.
AI systems trained on historical hiring data can replicate and entrench existing patterns of discrimination, producing biased outputs even where no discriminatory intent is present. Under German law, the origin of the bias does not diminish the employer’s liability. International employers should conduct a bias audit of any AI recruitment tool before deploying it in Germany and should be in a position to demonstrate that the tool does not produce discriminatory outcomes across protected characteristics.
Practical Steps for International Employers
Before deploying any AI-driven tool in a German HR context, international employers should take the following steps. Legal review of the tool against the EU AI Act’s high-risk classification criteria and associated obligations is an essential starting point. A data protection impact assessment (DPIA) should be conducted under Article 35 GDPR, which is mandatory where processing is likely to result in high risk to individuals – a threshold that AI-driven HR systems will typically meet.
Processing activities must be mapped against Section 26 BDSG and Articles 9 and 22 GDPR to establish the lawful basis for each category of data processed. Where a works council exists, the co-determination process under Section 87(1) No. 6 BetrVG must be initiated and completed before the tool goes live. Finally, the tool itself should be audited for bias against the protected characteristics set out in the AGG.
Seeking Legal Advice
AI regulation in Germany is developing rapidly, and the obligations it creates for employers do not exist in isolation. They intersect with established employment law frameworks – works council co-determination rights, data protection requirements under the BDSG, and anti-discrimination obligations under the AGG – in ways that require joined-up legal advice rather than piecemeal compliance.
At Schlun & Elseven Rechtsanwälte, our AI law team works alongside our employment law specialists to advise international employers on the full compliance picture. We assist with legal assessment of AI systems under the EU AI Act, data protection impact assessments, works council negotiations, and AGG audit preparation. If you are introducing AI-driven tools into your German HR operations, we recommend seeking legal advice before deployment rather than after a problem arises.


