38 AI - Gen and tonic

The life sciences industry stands on the precipice of a transformation – driven by the exponential growth of data, AI adoption and the increasing demand for faster, more accurate healthcare solutions.

At the heart of this transformation is Generative AI (GenAI). Unlike other forms of AI – such as numerical models machine learning or natural language processing – which primarily focus on tasks like classification or prediction, generative models like GANs (Generative Adversarial Networks) and transformers are capable of creating entirely new content.

By learning from existing data sets, they can generate everything from synthetic patient data to novel drug compounds.

Perhaps the most headline-worthy use of GenAI is in drug discovery. AI-generated molecules can be tailored for specific therapeutic targets, potentially reducing development timelines and uncovering candidates with improved efficacy or fewer side effects.

GenAI enables virtual simulations of how drug molecules might behave in the body, helping researchers identify the most promising candidates before moving to lab testing.

Another game-changer is synthetic data generation. In situations where patient privacy must be preserved or real-world data is scarce – such as with rare diseases – GenAI fills critical gaps by mimicking biological data sets. This not only enables training of predictive models but also allows broader participation in research and development across institutions and geographies.

Meanwhile in genomics, GenAI helps simulate and analyse genetic sequences, offering insights into gene function and enhancing the accuracy of gene editing technologies. Likewise, it plays a vital role in protein structure prediction, as seen with tools like AlphaFold, which are revolutionising our understanding of diseases at the molecular level.

Even diagnostics are seeing disruption. By generating synthetic biological samples or recognising new data patterns, GenAI is enhancing the development of AI-powered diagnostic tools, particularly for use in underserved or resource-limited regions.

Despite these breakthroughs, many AI and analytics projects struggle to move beyond proof of concept. Organisations face the challenge of scaling these innovations to production environments where they can deliver actionable insights.

This requires more than just technical integration; it demands a unified decisioning platform that can automate, govern and deploy decisions across the entire AI life cycle.

Key to this transition is embedding business rules and AI models into real-time workflows. By doing so, companies can deploy decisions at scale while maintaining trusted results and regulatory compliance. This also includes enabling collaboration between data scientists, business analysts and other stakeholders via intuitive tools and transparent version tracking.

While GenAI excels at accelerating research, it’s not about replacing humans; it’s about empowering them. The 80-20 rule applies here: GenAI can manage 80% of a task, freeing experts to focus on the remaining 20% that requires human judgement and oversight.

Moving to a 90-10 ratio, where GenAI manages 90% of a task, would dramatically challenge overall ROI. To realise this synergy between GenAI and humans, upskilling and training employees on topics like prompt engineering bias mitigation and data governance will be critical.

As the EU AI Act and guidance on AI and machine learning emerges from regulators like the FDA and EMA, AI in life sciences is under intense scrutiny. However, this also creates a unique opportunity for regulators and innovators to reshape frameworks for evaluation submission and compliance – always keeping patient safety and responsible use at the forefront.

Regulatory compliance and trust are foundational requirements. Black box AI would lead to a failure. Highly collaborative organisations should choose technologies that are understood and trusted, prioritising governance, versioning, traceability, explainability and auditing.

The promise of GenAI in life sciences is clear: faster discovery; deeper insights and more personalised care. But unlocking its full potential requires more than innovation; it requires strategic investment in people, platforms and partnerships. As organisations strive to improve ROI and scale responsibly, GenAI will not only shape the future of healthcare, it will define it.

Gen and tonic

Concept to reality