REMEDIUMREMEDIUM1561-59362658-3534Joint-Stock Company Chicot183010.32687/1561-5936-2025-29-4-389-396Original ArticleDevelopment of an ai system for auditing documents of quality management systems in the pharmaceutical industryKoshechkinKonstantin A.k.koshechkin@lpt.digitalSpichakIrina V.spichak@gxp-academy.orgKalashnikovRuslan V.rk@pharmabrain.ruZhirovaI. Vasilyevnazhirova@bsuedu.ruEurasian Academy of Good Practices, Moscow, RussiaFarMozg Limited Liability Company, Moscow, RussiaBelgorod State National Research University, Belgorod, Russia15122025438939615122025Copyright © 2025,2025In the context of strict regulatory requirements and the desire for continuous improvement, pharmaceutical companies are faced with the need to improve the efficiency and reliability of their quality management systems (QMS). Traditional approaches to the audit of QMS documentation, based on manual analysis, are time-consuming, subject to the human factor and do not always provide the necessary depth of analysis FarBrain, which uses artificial intelligence (AI) technologies for automated audit of QMS documents. The system is designed to analyze various types of documents, including standard operating procedures (SOPs), corrective and preventive action plans (CAPAs), quality policies and reports, for compliance with a wide range of international and national standards, such as EAEU GMP, ICH Q10, ISO 9001, PIC/S Guide, GAMP 5 and FDA 21 CFR Part 11. The system is based on natural language processing (NLP) and machine learning (ML) algorithms that automatically extract key information, identify inconsistencies and potential risks, and make recommendations for improvement. The article describes in detail the methods behind the operation of the system, including the architecture, the AI algorithms used, and the validation process in accordance with the principles of GMP. In addition, the results demonstrating the functionality of the system are presented and compared with other existing AI solutions in the field of quality management. The advantages, challenges and prospects of introducing AI into QMS audit processes are discussed, emphasizing the importance of integration with other digital systems and the development of predictive analytics to ensure a high level of quality and safety of pharmaceutical products.quality management systemartificial intelligenceinformation systemspharmaceutical quality systemсистема менеджмента качестваискусственный интеллектинформационные системыфармацевтическая система качества1.FDA. Guidance for industry. Quality systems approach to pharmaceutical CGMP regulations. Rockville; 2006. 36 p.2.ICH Harmonized Guideline Q10. Pharmaceutical Quality System. Step 4 version. — International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use, 2008. 16 p.3.FDA. Artificial INTELLIGENCE/MACHINE LEARNing (AI/ML)-Based Software as a Medical Device (SaMD) action plan. Silver Spring; 2021. 10 p.m.4.EMA. Refection paper on the use of artificial intelligence in medicine development. Amsterdam; 2023.25.GAMP 5: a risk-based approach to compliant GxP computerized systems. 2nd ed. Tampa; 2022. 298 p.6.ISO 9001:2015. Quality Management Systems — Requirements. Geneva; 2015. 29 p.7.Eurasian Economic Union. Good Manufacturing Practice for Medicinal Products: Appendix 1 to the EAEU GMP Rules (as amended by EEC Council Decision No. 90 of November 3, 2017). Moscow; 2017. 88 p.8.PIC/S Guide to Good Manufacturing Practice for Medicinal Products. PE 009—16 (Part I and Part II). Geneva; 2022. 352 p.9.EU Guidelines to Good Manufacturing Practice. Medicinal Products for Human and Veterinary Use. Annex 11: Computerised Systems. Brussels; 2011. 10 p.10.Code of Federal Regulations. Title 21, Part 11: Electronic Records; Electronic Signatures. Washington; 2023. 24 p.11.Reimers N., Gurevych I. Sentence-BERT: sentence embeddings using Siamese BERT-Networks // Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing. Hong Kong; 2019. P. 3980—3990.12.ISO/IEC 23053:2022. Framework for AI systems using ML. Geneva; 2022. 42 p.13.GAMP Good practice guide. Validation of AI and ML applications. Tampa; 2021. 112 p.