Validation of AI systems
Harmonising compliance and innovation in the use of artificial intelligence

GxP-compliant validation of AI systems
The introduction of artificial intelligence (AI) opens up enormous opportunities for companies to optimise processes, promote innovation and secure competitive advantages. Particularly in the pharmaceutical and life sciences industry, which is characterised by data-intensive processes, AI can revolutionise the development of new therapies, the optimisation of production processes and the efficiency of clinical trials.
However, the use of AI in regulated environments also brings challenges: strict requirements for safety, quality and transparency as well as new regulatory requirements such as the EU AI Act make the validation of AI systems indispensable. But how does the validation of AI systems differ from the validation of previous computer-aided systems?
Why is validation of AI systems so important?
Compared to traditional computerised systems with their rigid and deterministic algorithms, AI systems are more dynamic, data-driven and probabilistic. These characteristics present companies with new challenges in terms of implementation and validation:
Data quality:
Bias or poor training data can affect the performance of an AI system.
Data:
In addition to the quality of the data, separation of training, testing and validation data is also critical to ensure data integrity and system performance.
Explainability:
Decisions and predictions made by AI systems must be explainable – a key criterion in regulated industries.
Continuous monitoring:
Since AI systems can react sensitively to new inputs (model drifting), continuous performance monitoring is required over the entire life cycle.
Our approach to validating AI systems
DHC supports you in the GxP-compliant validation of your AI systems by combining proven methods of classic validation such as our own best-practice approaches and those of GAMP5® with specific approaches for AI.We take into account the special requirements of regulated industries and the latest regulatory developments, such as the EU AI Act. You can find more information on the EU AI Act here.
Would you like to find out more about the GxP-compliant validation of AI systems?
Then request our free white paper now and find out more about
- Potentials and challenges when using AI systems in a regulated environment
- Differences to the validation of traditional computer-aided systems
- Requirements due to regulations such as the EU AI Act
DHC CONSULTING SERVICES FOR THE VALIDATION OF AI SYSTEMS
- Initial maturity check:
Ranking of the maturity level with our specially developed AI scorecard - Strategic alignment:
Clarification of the goals and areas of application for AI in the corporate strategy. - Data strategy:
Analysis of workflows, data flows, data sources and data management.
- Governance and data sovereignty:
Clarification of authorisation concepts and copyrights. - Infrastructure:
Consultancy for digital infrastructure requirements. - Corporate culture:
Assessment of the corporate culture with regard to openness to innovation and expertise in the collection and qualification of data.
- Create clarity:
Which AI systems, use cases and business processes should be validated? - Understand regulatory requirements:
Which guidelines (GxP, EU AI Act, FDA, ISO standards, etc.) are relevant? - Carry out a risk assessment:
Which critical areas need special consideration?
- Define roles and responsibilities:
Who is involved in the validation process, who is responsible for what? - Basis for a roadmap:
Define action steps, resource requirements and time planning.
- Risk-based validation:
We analyse and assess the risks of your AI system and develop validation strategies based on this. - Data validation:
We check the quality, completeness and impartiality of your training and test data.
- Documentation and traceability:
From the definition of user requirements to the creation of the validation report, we ensure that all regulatory requirements are met. - Continuous monitoring & framework:
We help you implement processes and SOPs to regularly monitor and adjust your AI systems.
Expertise and GxP know-how
We have decades of experience in the validation of computerised systems in regulated industries.

Customised solutions
We offer customised validation strategies that take your specific requirements into account.

Compliance and innovation
We support you in using AI technologies safely and in compliance with regulations – without restricting innovation potential.
