As pharmaceutical companies race to adopt artificial intelligence across discovery, development and manufacturing, industry experts say the decisive factor will be the quality and readiness of data rather than the sophistication of algorithms.
Data readiness, not volume
Benchmarking by industry network BioPhorum, covering its members’ activity through 2025, found that data readiness, meaning access, availability, formatting and lineage, is the single largest barrier to AI deployment across the sector. Without clean, structured data as a starting point, meaningful validation of advanced AI models is simply impossible.
Participants in a drug-discovery industry discussion reached a similar conclusion: the challenge is not how much data companies hold, but whether it has been curated, contextualized and aligned with specific scientific questions.
A people and culture challenge
A Pistoia Alliance survey reported in January 2026 pointed out that pharma companies typically adapt existing AI technologies rather than invent new ones. The real work, it said, lies in data practices, organizational culture, workforce skills and management incentives. Scientists often prefer presenting conclusions in reports to sharing raw data, even as machine learning needs data in machine-readable formats.
A mid-year review of 2026 industry predictions echoed the theme: AI tools have arrived, but organizational readiness, data governance and regulatory gaps now limit their real-world impact.
Where AI is delivering
Early successes keep human experts firmly in charge. Companies and research organizations describe AI speeding up hypothesis generation while scientists retain control of validation, interpretation and strategic decisions. More advanced uses, such as predictive analytics and in silico modeling, are emerging and moving trial-and-error experimentation from physical laboratories into virtual simulations.
What it means
Data and analytics providers argue that fragmented sources and inconsistent standards are problems no algorithm can overcome, and that AI-ready data is now essential. As such firms also sell data services, their views are best read alongside independent surveys. For pharma leaders, the message is consistent: investment in data foundations and governance may matter more than the next model.
Written by: Bimbita Kar
Graphics by: Pramit Hazra
