Pharma GCCs tap artificial intelligence to fast-track drug discovery

They are not just cutting down drug development time, but also the cost involved in it. AI is reshaping how drugs and therapies are developed

Pharmaceutical global capability centres (GCCs) in India are emerging as critical hubs in the global effort to cut drug development timelines and costs, thanks to the accelerated adoption of artificial intelligence (AI) and generative AI (GenAI).

 

From early-stage discovery to clinical trial optimisation and regulatory processes, GCCs are co-creating AI solutions that are reshaping how drugs and therapies are developed.

 

“The process of ideation to getting a medicine in a patient’s hand is a 10 to 14-year process, but with artificial intelligence (AI), it will be reduced by 50 per cent. We should be able to see a higher success rate,” said Naveen Gullapalli, managing director of Amgen India, at a Nasscom event in April.

 

GCCs are deploying AI tools to accelerate molecule identification, optimise clinical trials, and build data-driven platforms for personalised medicine. The goal: faster, more precise, and cost-effective drug development, with billions in savings across the pharma value chain.

 

Generative models are now simulating billions of compounds, optimising for efficacy and safety, and cutting early-phase R&D timelines by up to 70 per cent. GCCs in India are co-developing these platforms and increasingly co-owning IP around molecular design and biomarker discovery. 

 

Beyond discovery, GCCs are integrating genomic, phenotypic, and real-world data to build pharmacogenomic models, accelerating personalised medicine pipelines. Clinical decision support tools built out of India are improving diagnostics across oncology, neurology, and beyond.

 

Karthik Padmanabhan, managing partner, GCC, at Zinnov, points to drug repurposing as another emerging area. “GCCs are using AI models trained on real-world evidence and longitudinal EHR data to identify new indications for existing drugs -- a strategy increasingly adopted in oncology and rare disease treatment pipelines. These models not only accelerate development timelines but also reduce R&D risk,” he said.

 

Clinical trials have always been the costliest and most time consuming part of any drug discovery process and the pharmaceutical GCCs are making an attempt to reduce it by predicting adverse events and refining patient cohorts using real-world and genomic data with AI tools developed in these centres.

 

US drug maker Amgen, which opened its technology and innovation site in Hyderabad last year, says it sees a huge potential to use AI across the whole chain of life cycle of innovative medicines. 

 

“It took several months of trial in a laboratory to determine the right molecules even two years back. Today, with machine learning (ML), we can eliminate hundreds of molecules that do not hit the right viscosity range which is currently 80 per cent accurate. This will only get better as a solid use case to find the right molecule,” Gullapalli added.

 

Amgen announced earlier this year its plans to invest $200 million in this centre that is focused on increasing use of AI and data science to support development of new medicines.

 

Swiss pharmaceutical company Novartis uses AI in four areas. They are clinical trial design, investigator and site selection and patient enrolment, clinical operations, and clinical document generation. For clinical trials, it encompasses streamlining protocol development to more efficiently identify study sites, enrolling diverse patient populations or generating complex clinical documents more accurately and efficiently.

 

Novartis has developed its own Protocol.AI, a technology platform, that helps teams develop better concept sheets and protocols for clinical trials. By using natural language processing (NLP) and advanced Gen AI techniques, the tool extracts and maps protocol information from a wealth of external and internal knowledge, including real-world evidence and its own library of concept sheet and protocol writing.

 

For investigator and site detection and patient enrolment, the Swiss company has developed Clinical Intelligence Platform (CLIP), a foundational tool which uses its AI engine by pinpointing the countries, trial sites, and investigators most likely to enroll the target patient population for a given trial.

 

Sadhna Joglekar, head of the India hub, said that for clinical documentation, it has tied up with Yesop, an AI-powered platform that helps life sciences team tackle complex documentation.

 

“That allows our teams to focus their time and energy on critical reviews and quality assurance. Initial feedback from medical writers indicates improved quality of content and a 30 per cent reduction in authoring time,” she added.

 

The Novartis Corporation Centre (NOCC) in India, which houses about 2,800 people out of the 12,000 in the development organisation globally, has contributed to over 150 projects and currently supports 51 global clinical trials involving 2,400 patients.

 

Novo Nordisk, the Danish company behind the blockbuster weight loss drug Wegovy, said that its India technology centre works on two areas, clinical reporting and global regulatory affairs. That involves volumes of paper work and quality checking which are manual and laborious.

 

“We partnered with a start-up in Bengaluru to get a tool that does all the quality checks as it is automated with a final validation with a human before submission. The checking time has reduced to 40 minutes from 70 hours previously,” said John Dawber, corporate vice president and managing director of Novo Nordisk Global Business Services.

 

“In drug development, it has been an academic hypothesis to do molecular modelling for decades. This means you take the molecule, put it inside the lab, and test it to decide if it’s a drug candidate. That still happens but with ML and AI, you can do the pre-laboratory work much more effectively. There is a greater chance of success with computer modelling,” Dawber added.

 

For Sanofi, it involves using ‘plai’, an AI solution developed with Aily Labs which leverages more than 1 billion data points across the French drug maker from clinical trial design to portfolio management. The company’s target discovery engines have delivered seven novel drug targets in just a year, while its mRNA modelling solution has cut mRNA design time in half. 

 

“Our integrative clinical data (ICD) team, working with clinical real-world evidence, is harnessing machine learning to process real-world data (RWD) to improve the efficacy of our trials and to identify new indications for existing assets in development. Our clinical operations teams are using plai during the recruitment phase of trials to make sure our studies are sufficiently diverse, providing the data we need to develop treatments that benefit as many patients as possible,” the company said in an emailed response.

 

“This isn’t just about adoption, India’s GCCs are reshaping global pharma’s AI playbook by scaling and localising it,” Padmanabhan said.