A featured contribution from Leadership Perspectives: a curated forum reserved for leaders nominated by our subscribers and vetted by the Construction Tech Review Advisory Board.
CSL Innovation GMBH
Trusted Data Foundations for Pharma R&D
Axel Dietrich
Axel Dietrich, Global Head of Data Excellence at CSL Innovation GmbH, champions a trusted data culture, treating data as capital. He unlocks legacy R&D data with semantic layers, advances AI and automation, and builds enabling governance. He prioritizes explainable, validated, compliant systems and partners with decision-makers to prove ROI and accelerate outcomes.
In this interview, Dietrich explains how trusted data foundations, semantic context, and responsible AI transform pharma into a decision-ready, compliant enterprise. His focus on governance as an enabler unlocks legacy knowledge, proves ROI, and moves leaders beyond pilots to measurable, faster outcomes.
Building a Unified, Trusted Data Culture
I see data culture as a continuous effort that starts with understanding people’s realities, what they are trying to achieve and what gets in their way. At operational levels, I focus on day-to-day issues. At strategic levels, I align with long-term objectives and the regulatory context. My first step is always to meet teams where they are and earn trust by showing tangible outcomes.
Trust grows when I choose the right use cases, problems that visibly save time, improve quality, or remove friction. If I can demonstrate a clear return on investment, I win support from decision-makers. I work closely with those “deciders,” identifying sponsors eager to back the work, and then prove, with results, that data, automation, and AI deliver better reports, products, and faster cycles.
Governance enables this culture; it does not control it. I define clear accountabilities and responsibilities across levels and make the case that governance exists to give the business better access to better data. Executives must buy in, owners must be identified, and domains must be made connectable so knowledge compounds. When governance is positioned as an enabler, adoption follows.
This cultural foundation is now reflected in our long-term organizational objectives. We delivered a data strategy that explains how we will enable data, AI, and automation for success, and we tied it to the increasing expectations from regulators like
the FDA and EMA for digital access to data and documentation. Embedding this into objectives clarifies priorities and accelerates alignment across functions.
From Automation to Intelligence
Pharma is entering a new phase. The shift is from process automation to machine intelligence that understands scientific language and connects research data into real-time knowledge. The question is not whether AI will reshape discovery but who will lead with trusted, intelligent data.
Curiosity, agility, and trust are not slogans. They are habits built by taking on unfamiliar problems and delivering outcomes that matter.
Our greatest untapped potential lies in legacy data—vast, valuable, and often hidden in silos across the enterprise. Much of it mixes structured and unstructured forms, locked behind specialized terminology and incompatible systems. Modern AI, including generative and agentic approaches, can navigate this landscape, link domains and surface the enterprise’s collective knowledge at speed.
To realize that potential, data readiness is pivotal. High-quality, contextualized data with a semantic layer is the foundation; without it, even the most sophisticated algorithms remain superficial. Building that semantic layer is not an IT project—it requires the business and scientific owners to define meaning, connections, and decision pathways so AI can truly “speak” the language of science.
When we treat data as capital—complete with quality, lineage, and semantics—we enable continuous learning systems where insights accumulate over time. Neural networks trained on scientific data can already uncover patterns beyond human cognition, supporting target identification, screening and interpretation. The combination of human expertise and machine reasoning is accelerating discovery and redefining pace.
Governance, Compliance, and Leadership for Responsible AI
In regulated environments, trustworthy AI must be validated, explainable and compliant by design. I emphasize auditable, ethical and transparent systems that meet regulatory expectations while enabling speed. Trust, transparency, and governance become differentiators, turning restrictions into a framework for responsible acceleration rather than barriers to progress.
Effective governance clarifies who owns which data, how domains interconnect and how responsibilities cascade. Done well, it enables access to better data and supports cross-domain linkage so enterprise knowledge can be reused and scaled. Executive sponsorship is essential, and if sponsorship is missing, frameworks remain theoretical and adoption stalls.
Leadership must move beyond pilots and experimentation to build trusted data ecosystems. The future of digital transformation is not technology adoption alone. It is about creating enterprises that think, with systems that learn from every experiment, connect every dataset and surface knowledge at the speed of thought. Those who master semantics and trusted foundations will define the next era of discovery and patient treatment.
For those entering this field, my guidance is simple. Change domains and be willing to feel like an apprentice again. New perspectives fuel learning curves and prepare you for constant technological change. Curiosity, agility, and trust are not slogans. They are habits built by taking on unfamiliar problems and delivering outcomes that matter.
