AI Readiness in Biopharma: Overcoming Infrastructure, Talent, and Data Challenges

Bridging the AI Gap in Biopharma

Artificial intelligence (AI) and machine learning (ML) are widely recognized as transformative forces in biopharma, yet few companies are fully prepared to harness their potential. According to Benchling’s latest report, only 14% of large biopharma and just 3% of small companies are AI-ready. The question remains: what is holding the industry back, and how can IT leaders accelerate AI adoption?

In this Pharma IQ Q&A, we speak with Stephen Deasy, Chief Technology Officer at Benchling, to explore the biggest AI readiness challenges in biopharma, including data infrastructure, wet-dry lab integration, SaaS adoption, and FAIR data principles. He also shares practical strategies for attracting top AI/ML talent, optimizing data management, and future-proofing IT investments.

AI Readiness Gaps: Why Is Biopharma Behind?

Pharma IQ: The report highlights that only 14% of large biopharma and 3% of small companies are AI-ready. From an IT perspective, what are the most significant challenges in bridging this readiness gap?

Stephen Deasy: For large biopharma, the main hurdle with AI readiness is complexity. They’re managing massive amounts of data on systems that were never built for today’s AI-driven world. Imagine trying to integrate and automate workflows while juggling over 20 different software tools. That’s what large companies are dealing with — 43% of IT execs at these companies are supporting more than 20 different software applications. Instead of setting the stage for AI, they’re busy keeping the lights on, maintaining and securing all those systems.

Smaller companies face a different kind of struggle. They’re focused on the foundation — getting systems like R&D platforms in place. AI can seem like a distraction when they’re still working on building the infrastructure they need to even consider it.

And for both large and small biopharma, one of the biggest challenges is connecting workflows. Wet lab experiments and dry lab data analysis often feel like separate worlds, but they need to work together seamlessly if AI is going to deliver real insights. It’s not easy, but it’s crucial for moving the whole industry forward.

Pharma IQ: Access to skilled AI/ML talent is a major barrier, especially for small biopharma. What strategies can IT teams employ to attract and retain the right talent for AI initiatives?

Stephen Deasy: For small biopharma in particular, competing with big tech companies for AI talent can feel like an uphill battle. But there’s a big advantage here: impact and purpose.

Biopharma isn’t just about building tech. IT teams can attract talent by emphasising the life-changing impact of their work in biopharma — like using AI to revolutionise drug discovery or automating workflows to accelerate the time to market. Tech talent, whether engineers, data scientists, or systems architects, are motivated by seeing the impact of their work, and this is as meaningful as it gets.

Speaking from personal experience, I’ve built my career in technology, working at fast-growing tech companies like Atlassian, VMware, and EMC. What motivated me to get into biotech with Benchling was hands-down the mission and opportunity for impact.

Once you’ve brought tech talent on board, keep them engaged by creating an environment where they can explore, experiment, and solve real problems. Also, break down silos between IT, R&D, and other teams. When workflows are streamlined, and everyone’s aligned, your team spends less time on maintenance and more on impactful, rewarding projects.

Pharma IQ: The report shows IT leaders are more optimistic than R&D about AI/ML's impact on speed, quality, and scalability. How can IT work with R&D to align priorities and demonstrate AI’s value?

Stephen Deasy: Alignment starts with speaking the same language. IT teams need to show R&D how AI directly supports their goals — whether that’s speeding up timelines, improving data quality, or scaling processes.

But don’t just talk about potential. Show results. Use real examples where AI has solved a specific bottleneck or improved an R&D workflow. If you can put numbers behind it — like time saved or costs reduced — even better.

Pharma IQ: Many large biopharma companies support custom-built software applications. How do you balance the flexibility of custom solutions with the scalability of off-the-shelf tools?

Stephen Deasy: Supporting custom software at scale can be a massive logistical burden. Imagine managing over 20 applications — each with its own integrations, updates, and security needs.

While custom-built tools offer flexibility, they often slow you down in the long run. Collaboration suffers when systems don’t talk to each other, and adding new capabilities becomes a mountain to climb.

Off-the-shelf tools, on the other hand, can scale more easily but may require foundational changes first. The key is knowing when to go custom and when to standardise. Review your existing tools. Which ones are truly essential? Consolidate the rest onto scalable, shared platforms that free up resources and simplify workflows.

Pharma IQ: Large biopharma sees wet and dry lab integration as a major challenge. What role does IT play in connecting these environments, and how critical is this for scaling AI/ML?

Stephen Deasy: Working across different biopharma companies, the largest hurdle I see when it comes to AI preparedness is in uniting wet and dry lab workflows. Only 41% of large biopharma are prepared connecting wet and dry lab workflows.

IT plays an important part bridging this gap.That means implementing a structure where the data producers (wet labs) and data consumers (dry labs) are tightly connected. It involves having shared systems of record, common semantics, and seamless collaboration between experimental and computational scientists, and then scaling the wet lab to support testing of all the ideas generated in the dry lab.

But it’s not just about the tech. It’s about rethinking processes and fostering collaboration across teams. Many of the companies we work with are now embedding engineers and data scientists into wet lab teams from day one. 

Pharma IQ: SaaS adoption is slower in biopharma than in other industries. What steps can IT teams take to address data security and cultural resistance to accelerate adoption?

Stephen Deasy: SaaS adoption is lagging in biopharma — only 22% of small and 17% of large companies have the majority of their software in the cloud. Why? Concerns about data security and resistance to change are the main issues that IT needs to address.

To move forward, IT teams need to tackle these concerns head-on: Show how SaaS meets security requirements and aligns with the company’s mission. And work with SaaS vendors to demonstrate real-world examples of faster timelines, better collaboration, enhancements with data analysis, and streamlined workflows.

Address cultural resistance by reframing the conversation. Scientific SaaS isn’t just a new tool; it’s a way to accelerate progress toward your bigger goals while staying secure and compliant.

Pharma IQ: FAIR data principles are essential for effective AI/ML. How is IT addressing challenges like standardising metadata and ensuring data interoperability across diverse lab instruments and systems?

Stephen Deasy: At large biopharma companies, it’s common to work across over 100 lab instruments. These instruments, from cell culture systems to next generation sequencers, are often operating in isolation without automated data capture. This leaves R&D data locked in silos, requires manual data transfer, and definitely slows down the science.

The lack of standardised data formats is a big bottleneck. We hear this from nearly every biotech and biopharma company we work with, and it’s backed by our research: 61% of large biopharma and 51% of small biopharma report this as their top challenge when it comes to connecting lab instruments. Without universal standards, integrating lab instruments and making data flow seamlessly is a struggle. On top of this, resources are tight. Many teams lack the bandwidth or budget to build custom integrations to allow for flow of data with these instruments.

These gaps with instrument and data connectivity matter, they hold back AI/ML. Despite AI being a top investment priority, biopharma can’t fully leverage it without connected lab instruments and FAIR (findable, accessible, interoperable, and reusable) data.

The path forward is clear: industry-wide data standards to help get the critical data into the hands of scientists quickly, and smarter low-code or no-code integration tools.

Pharma IQ: With AI and data platforms dominating future investment priorities, what steps should IT leaders take now to build scalable, integrated infrastructures that can adapt to emerging technologies?

Stephen Deasy: Biopharma companies, big and small, are zeroing in on connectivity and orchestration. The priority is clear: automate and centralise the entire data flow, from creation to analysis. This streamlined approach is critical to meet the demand for high-throughput operations and to support the increasing role of AI and computational tools in R&D.

With this, the focus is shifting to adopting R&D data platforms, connecting lab instruments, and embracing SaaS solutions. While progress is being made, challenges still exist. Large biopharma must tackle fragmentation and data security concerns, while smaller companies face the task of finding solutions that deliver real value without straining their resources.

Some are already doing this well. For example, we worked with Sanofi to move away from fragmented legacy systems to a unified digital platform that is used by more than 30 global teams and 1,500 scientists. This means teams can now benefit from seamless data sharing, whatever team they work in — this centralized data platform is the foundation for the company’s ambitions of ‘AI at scale.’ Zealand Pharma has also moved the cloud with Benchling, in order to organise, analyse and share data more easily within the company and with its high amount of external contract manufacturing organizations (CMOs). This more streamlined data management helps Zealand to scale operations in its fast-growing space of obesity and metabolic disease.

Clearly, there is momentum building. By prioritising scalable, connected systems, IT leaders can pave the way for a future where AI drives innovation and breakthroughs happen faster than ever.

More Insights on AI in Biopharma

Pharma IQ continues to explore the latest AI advancements in life sciences. For more industry perspectives on how AI is transforming drug discovery, clinical trials, and pharmaceutical manufacturing? Join industry leaders at Pharma IQ’s AI for Pharma & Healthcare Summit 2025 to explore real-world AI applications and strategies for accelerating innovation in life sciences.