AI Transparency & Responsible Use Policy

Last updated: 23 April 2026

ClausePilot Korlátolt Felelősségű Társaság (registration number: Cg.01-09-454891; registered seat: 1163 Budapest, Karát utca 33., Hungary) ("ClausePilot", "we", "our") is committed to building trust in the use of artificial intelligence (AI). This policy explains how AI is integrated into our platform, the safeguards we apply, and the responsibilities of our users.

1. Purpose of AI in ClausePilot

ClausePilot integrates document drafting, legal research, document analysis, and secure document management in a unified platform. Unlike generic AI tools, ClausePilot is purpose-built for administrative and legal work:

Minimized risk of hallucination in legal research, powered by advanced RAG (Retrieval-Augmented Generation) technology that grounds all outputs in verified legal sources to ensure high factual accuracy
Multi-capability workflows covering drafting, analysis, research, and secure document vault management
AI designed as a supportive tool that enhances professional judgment rather than replacing it
Users retain full responsibility for reviewing, validating, and applying outputs in their work

2. No Legal Advice

ClausePilot does not provide legal advice.

Any AI-generated text, clause, or analysis is a drafting aid only
The ultimate responsibility for legal accuracy, compliance, and client advice lies with the user
All outputs require professional review and validation before use

3. Transparency on AI Technology & Sources

Technology Stack

ClausePilot combines proprietary legal knowledge curation with state-of-the-art large language models with zero data retention obligations while data is hosted in secure EU data centers (AWS Europe)
Our RAG-based vault technology ensures AI responses are grounded in verified legal documents and precedents from your firm's knowledge base
Advanced retrieval systems prevent AI hallucinations by anchoring outputs to authoritative legal sources

Data Practices

We do not train AI systems on client data
None of the third-party AI providers train their system on our clients’ data or the content generated by our clients.
Strict zero data retention obligations for all third-party AI providers. All User-Generated Content is stored under the user’s exclusive control within the platform's secure EU data centers (AWS Europe).
Models are continually tested and improved to reduce risks of inaccuracy, bias, and hallucination

4. Safeguards and Human Oversight

Security & Privacy

All data is processed under strict confidentiality, GDPR compliance, and professional secrecy standards
Enterprise-grade security infrastructure with data encrypted at rest (AES-256) and in transit (TLS 1.3)
Hosted on secure EU data centers (AWS Europe) with industry-standard security protocols
Users maintain full control over data retention and deletion
Where AI outputs are processed by U.S. API providers (Appendix 2 of the Data Processing Agreement), transfers occur under SCCs/DPF; the primary stored version of privileged documents remains in EEA data centres at all times. The content submitted for analysis is immediately purged by the API provider after the processing is complete and is not stored in the U.S.
For detailed cookie practices, see our Cookie Policy

Quality Assurance

Human-in-the-loop validation: All AI outputs are designed for review and adaptation by qualified professionals
Continuous accuracy testing and bias monitoring protocols
Source attribution and citation tracking for all AI-generated content

5. EU AI Act Compliance & Future-Proofing

ClausePilot is designed to align with the principles and upcoming requirements of the EU AI Act.

Current Measures

Clear labeling and identification of all AI-generated content
Comprehensive risk assessment and management protocols
Quality management systems with human oversight requirements
Detailed logging and audit trails for AI decision-making processes

Ongoing Compliance

Regular accuracy and bias testing procedures
Transparency reporting on AI system performance
User notification systems for AI limitations and appropriate use cases
Continuous monitoring and improvement protocols

6. Responsible Use by Users

To maintain platform integrity and legal compliance, users must:

Use ClausePilot only for lawful and professional purposes
Never rely on AI outputs as legal advice
Apply professional expertise and judgment before sharing AI-generated content with clients or third parties
Maintain awareness of AI limitations and appropriate use cases
Report any concerns about AI accuracy or inappropriate outputs

7. Continuous Improvement & User Feedback

ClausePilot is committed to ongoing enhancement:

Regular model updates based on latest legal AI research and user feedback
Transparent communication about system updates and new capabilities
User feedback integration into product development cycles
Industry collaboration to advance responsible AI practices in legal technology

Our Commitment

ClausePilot is built for lawyers, by lawyers. Our mission is to empower legal professionals with cutting-edge AI technology while respecting the core values of legal practice: independence, professional responsibility, and confidentiality. We believe AI should enhance, not replace, the expertise and judgment that define exceptional legal work.