GLP-1 giant Novo Nordisk partners with OpenAI as pharma industry’s AI race accelerates Proactive uses images sourced from Shutterstock
The Danish drugmaker becomes the latest major pharmaceutical company to embed AI across its entire drug development pipeline, from discovery to commercial operations.
Novo Nordisk (NYSE:NVO), the Danish pharmaceutical giant best known for its diabetes and obesity treatments, has formed a partnership with OpenAI that will apply the artificial intelligence company’s most advanced capabilities.
It will deploy the technology across its global operations, from early drug discovery through to manufacturing, supply chain and commercial functions.
The partnership will use AI to analyse complex datasets, identify drug candidates and reduce the time required to move from research to patients, with pilot programmes launching across research and development, manufacturing and commercial operations and full integration targeted by the end of 2026.
OpenAI will also assist Novo Nordisk in upskilling its workforce and improving AI literacy across the organisation.
Mike Doustdar, president and chief executive of Novo Nordisk, said the partnership would allow the company to analyse datasets at a scale previously impossible and test hypotheses faster than ever, with the goal of discovering new therapies and bringing them to market more quickly.
Sam Altman, chief executive of OpenAI, said the collaboration would help Novo Nordisk accelerate scientific discovery, run smarter global operations and redefine the future of patient care.
The tie-up reflects a broader and accelerating shift across the pharmaceutical industry, where AI is moving from isolated experiments to core infrastructure embedded throughout the research and development process.
Novartis has described using generative AI to design 15 million potential drug compounds computationally before narrowing laboratory work to around 60 molecules, dramatically compressing what was previously a far longer process.
The Swiss drugmaker has also used AI-driven simulations to identify gene candidates for rare kidney disease.
This is a task, it said, would have been prohibitively slow without the technology and has a collaboration with Isomorphic Labs, the AI drug discovery company spun out of Google’s DeepMind.
French pharmaceutical group Servier has signed a strategic partnership with AI drug design firm Iktos covering more than one billion euros in potential multi-target drug discovery programmes across oncology and neurology.
Takeda, the Japanese drugmaker, has partnered with Arrayo to use machine learning to pre-screen antibody candidates for stability, affinity and manufacturability, accelerating a design process that previously relied heavily on manual chemical intuition.
Industry data compiled for 2026 suggests half of drug developers using AI already report faster time-to-target identification, while 42% report improved accuracy and hit rates in their scientific models.
Despite the momentum, the technology has yet to produce a drug that has achieved regulatory approval in its own right.
AI can compress early discovery timelines by an estimated 30% to 40% and reduce the time required to identify a preclinical candidate from three to four years to between 13 and 18 months.
However, clinical trial durations and regulatory review timelines remain bound by biology and patient enrolment, neither of which AI can shortcut.
The Novo’s partnership is structured with strict data protection, governance and human oversight provisions, reflecting a wider industry push to make AI deployment defensible to regulators as well as scientists.
The US Food and Drug Administration published draft guidance in January last year setting out a risk-based framework for assessing AI models used in regulatory decision-making, a development analysts say is beginning to reshape how pharmaceutical companies design their AI systems from the outset.