The hardest part is not getting started, but maintaining continuous AI operations.
Pharma IT Insights
AI in Life Sciences: When Expectations Meet Reality
By Nick Larsen
September 19, 2025
Pharma IT Insights
AI in Life Sciences: When Expectations Meet Reality
By Nick Larsen
September 19, 2025
There’s no doubt among the leaders in life science companies: artificial intelligence is high on the agenda. Headlines speak of “revolution”, “acceleration” and “miracle solutions”, but behind the shine lies a more down-to-earth reality.
Many companies have realized that AI is not a switch you can simply turn on. It requires leadership, data discipline, and structured collaboration; otherwise, expectations risk colliding with reality.
Authorities and internal quality control departments require a transparent and well-documented process on the use of AI. Delivering results is not enough; it is necessary to explain how those results were achieved. In practice, this means that AI-projects must be developed the same way like any other project in life sciences with transparency, traceability, and “compliance by design.” These are not additional costs, but fundamental prerequisites.
The pressure of expectation, however, often comes from top management. They want to see AI in action, drive innovation, and communicate stories of transformation to investors and business partners.
Meanwhile, project teams struggle with unstructured and imprecise data, siloed systems, and a lack of standards and metadata, which makes it difficult to consolidate knowledge and create a consistent foundation for AI. On such a basis, AI does not create value. On the contrary, models can be trained with bias and errors, which undermines value and increases risk.
This is where data quality becomes the invisible yet crucial foundation. It is not about quantities, but about structure: if data is incomplete, fragmented or lacks ownership and documentation, AI initiatives quickly become expensive experiments that cannot be scaled, or may end in regulatory dead ends.
In the middle of the complexity lies a way forward. The companies that succeed in moving from “hype to discipline” will emerge stronger.
Three Clear Elements for Success
- Data governance with well-defined roles, standards, and quality control.
- Compliance by design, where AI solutions are developed with audit trails, validation and transparency from the beginning.
- Focus on realistic use cases, such as small, targeted projects which can be scaled rather than large complex visions.
All new technology presents challenges, especially in an industry where compliance is crucial. Many questions arise: Where do we start? What’s important? Who takes the lead?
A good starting point is:
- Start with the data foundation
Without proper data governance and architecture, progress is impossible. This work begins with standardization and ownership, extending to traceability and quality. - Define your AI-focus
Is the purpose to strengthen the core business, automate processes or provide tools for employees? Transparency separates noise from strategy and gives the projects a clear direction. - Take ownership in the management
AI is not merely an IT task. It is the management’s responsibility and opportunity to set direction, momentum and empower decision-making. This is not about hype, it is managerial decisions that drive tangible results.
From hype to operation:
Implementation is just the beginning; the focus must now be on ensuring data, models, and processes remain high-quality while continuously auditing performance, safety, GDPR, and compliance. This requires systems capable of monitoring operations and performance in real time.
It is not about smarter models, but about smarter processes and systems that make AI reliable in practice.
Existing processes, systems, and capabilities can quickly become barriers if they are not modernized. Therefore, companies should map technical debt, standardize workflows, and establish new roles and governance structures that support AI. The goal is to create an organization where innovation and regulation go hand in hand, and where AI becomes a catalyst for entirely new business models.
Exploiting AI’s full potential requires more than technology; it requires organizations to reinvent themselves. Existing structures, processes, and competencies are rarely designed to cope with the speed and complexity that AI introduces.
For strategic decision-makers, this means AI investments should be structured around thorough assessments rather than haste. Being a “first mover” is less important than being the “best mover”—which translates to be able to validate, document, and convince regulatory authorities. Companies that do this can achieve shorter time-to-market, stronger pipelines, and a safer journey from data to approved product.
A recent Danish study shows that seven out of ten employees use AI-tools without employer approval, while only 30% of companies have clear AI guidelines. This creates a gap where control is lost, and innovation risks being reduced to isolated initiatives rather than a collective business strategy.
To generate real value, the use of AI requires clear structures and a shared strategic direction. Without proper governance and well-defined processes, organizations face genuine risks related to data security, compliance, and the erosion of collective business capabilities. AI is not an IT project, but a management task focused on embedding AI into the company’s strategy, developing skills, and fostering a culture where innovation and regulation go hand in hand.
Without clear guidelines and shared understanding, organizations risk ending up with a fragmented mix of tools that are neither secure nor scalable.
Ultimately, this is how AI in life sciences moves from impressive visions to tangible business value. Compliance and data quality are no longer merely regulatory requirements, but they are clear competitive differentiators. Organizations that build AI initiatives on discipline, transparency, and targeted experimentation will be best positioned in a future market where innovation and regulation converge.
When management takes the lead while effectively engages employees, a synergy between innovation and discipline emerges. This is where AI transitions from hype to real business value, turning expectations into reality.
💡 Ready for compliant AI?
About the author
Nick Larsen is Pharma IT’s Director of Data, AI & Migrations. He leads the company’s efforts to transform complex data ecosystems into scalable, compliant, and AI-enabled solutions for the life sciences sector. Nick oversees GxP-compliant data migrations, validated pipelines, and integration projects across regulated environments. He brings deep expertise in bridging technical and business needs, helping clients adopt modern data strategies that support international growth. By combining hands-on execution with a clear focus on long-term value, he ensures Pharma IT delivers reliable, future-ready data solutions that meet both operational and compliance demands.
