EU AI Act
The EU Artificial Intelligence Act (Regulation (EU) 2024/1689) classifies AI systems according to the risk their use poses. MeetandGreet.AI is a generative AI system for interview practice; we provide it as the provider within the meaning of the Act. This page explains how we classify the system, which obligations we meet, and — importantly — where the limits of its intended purpose lie, in particular with regard to recruiting.
Risk classification
Risk classification under the AI Act depends on a system's intended purpose. Used as intended — self-directed interview practice by candidates preparing on their own initiative — MeetandGreet.AI is not a high-risk system: it generates natural-language questions and practice feedback in a conversational interface and is subject to the transparency obligations of Art. 50 AI Act. We are, however, well aware that AI systems used for the recruitment or selection of natural persons — for example to evaluate candidates in interviews — are classified as high-risk under Annex III No. 4 of the Act. Precisely for that reason, the intended purpose of our system expressly excludes such use; the section 'Use in recruiting' below explains what this means for companies.
Intended purpose
The AI system is provided solely for practice and training: candidates prepare for real interviews and receive a first, informal impression of their performance. Typical uses include job-interview preparation, assessment-centre practice, university admissions, language practice, communication and presentation skills, career preparation, and general conversation practice. The system is deliberately designed as a preparation and feedback tool. It is not intended to conduct, replace, or decide any part of a real recruiting, examination, or admissions process — that remains the job of human beings.
Use in recruiting — responsibilities of companies
AI systems used for the recruitment or selection of natural persons — in particular to screen or filter applications or to evaluate candidates in interviews — are high-risk AI systems under Annex III No. 4 AI Act. Any organisation that integrates MeetandGreet.AI into an actual recruiting, screening, or selection process therefore acts on its own responsibility as the deployer of a high-risk AI system and must itself comply with all obligations that apply to such use — including Art. 26 AI Act (appropriate human oversight, monitoring, information duties towards affected candidates), the GDPR, and applicable employment law. This applies equally to any results an inviting organisation receives through our platform. Our position is clear: MeetandGreet.AI exists to give candidates and organisations a first, non-binding feedback signal and a safe space to practise. It is not designed, validated, or approved for making or preparing hiring decisions, and it is not meant to replace human-led recruiting. Hiring decisions must never be based on scores or feedback generated by this system.
Transparency obligations
In line with Art. 50 AI Act, every interview is clearly labelled as an AI simulation. Each panelist is an AI persona — not a real person. AI-generated questions, AI-generated feedback, and AI-generated voice are disclosed in the UI before and during the session, so you always know that you are interacting with an AI system and that its output is machine-generated.
No real hiring or admissions decisions
The AI system does not make, and does not represent, decisions about job applications, hiring, admissions, examinations, promotions, or contract formation. Generated scores and feedback are illustrative practice output only — a first impression, not an official or binding assessment, and no measure of suitability for a specific position. Employers and other organisations must not use them as the basis for decisions about real applicants.
Limitations of the AI
AI outputs are probabilistic. They may be inaccurate, incomplete, biased, or unsuitable for a specific situation, and scores can vary between sessions even for comparable performance. The platform is not a substitute for legal, tax, medical, psychological, career, or HR advice. Do not rely solely on AI-generated feedback for important decisions.
No biometric processing
We do not perform emotion recognition, biometric categorisation, or any inference about your identity, demographics, or psychological state from your voice or video. Audio is used solely to transcribe what you said. Emotion recognition in the workplace is, moreover, a prohibited practice under Art. 5(1)(f) AI Act — one more reason such analyses have no place in this product.
Human oversight
You can review the full transcript and scores from any session, dispute the outcome via our contact address, or delete the record. AI feedback is illustrative practice — not a professional assessment — and you remain the sole decision-maker about how to use it. Where an organisation invited you to a session, that organisation is responsible for appropriate human oversight of any use it makes of the results.
Data governance
We do not train our own models on your interview content. Third-party providers (Google Gemini, Anthropic Claude, Google Cloud Speech) process inputs under their own enterprise terms with no training on customer data.
Ongoing improvement
AI models and system behaviour are updated periodically to improve quality, safety, and performance. As a result, the interviewer’s behaviour and the quality of its responses may change over time.
Last updated: July 14, 2026