Clarkston Consulting https://www.facebook.com/ClarkstonConsulting https://twitter.com/Clarkston_Inc https://www.linkedin.com/company/clarkston-consulting http://plus.google.com/112636148091952451172 https://www.youtube.com/user/ClarkstonInc
Skip to content

Opportunities for AI in Drug Product Development

Generative artificial intelligence (AI) is quickly expanding to all industries to drive efficiencies across business processes, now including the life sciences research and GCP environments. Now, the federal government has found a role for large language models (LLM) to enhance their workflows. 

The FDA announced on May 8, 2025, that the agency will implement artificial intelligence to all FDA centers by the end of June 2025. This decision is following a recently completed pilot program to use generative AI to assist in scientific reviews, which demonstrated the ability to drastically reduce the time required to complete tedious and repetitive tasks. 

Though business leaders within the life sciences industry have taken the opportunity to understand the landscape of AI, including infrastructure and use cases, some are hesitant to fully adopt this new platform due to compliance concerns and to adhere to regulatory standards.  

Now that the regulatory authority is embracing this novel technology, it’s time for companies to fully consider the potential benefits of these solutions while creating a roadmap to ensure holistic training data and validation. Just as the FDA is exploring ways to best use artificial intelligence, there is a breadth of opportunities for pharmaceutical companies to consider. In this piece, we unpack opportunities for AI in drug product development.

Use Cases for AI in Drug  Product Development 

The expansion of AI has developed anticipation due to its enablement of complex analytical techniques, such as the ability to use AI in drug development by screening chemical compounds. These novel applications have a real potential to create breakthroughs in the pharmaceutical development process, but require a strong data foundation to achieve success. To achieve these long-term strategic goals, consider starting with early use cases of AI while continuously developing your AI infrastructure, building from a crawl to a walk, then to a run. 

Crawl 

A starting point, or a crawl, when using generative AI should include a repetitive process with the potential to save valuable time. For example, a chatbot could be used to quickly educate the sales force team about the products and disease indication. This ChatGPT-like model could summarize marketing material or scientific articles in a way that is digestible and easy to understand. This would require a model that can integrate into a company’s document-sharing platform to connect directly to scientific and marketing documentation. This will be especially important if the FDA shortens time from BLA submission to approval by using the enhanced scientific review. 

Walk 

As an organization grows in their data and analytics capabilities, the group can transition to more complex use cases, a walk, including quickly analyzing large datasets. For example, an LLM could be connected to clinical trial datasets and automatically determine product efficacy on a study cohort. This more sophisticated application will require increased investment in data engineers to harmonize clinical data, data scientists to develop analytical tools, and technology platforms to store data. In addition to the technology, it’s necessary to increase investments to validate the tool and train users to know when to trust the output and when additional verification is required. 

Run 

Following the implementation of these initial use cases, leaders can prepare for greater investment to apply intelligence to all areas of the drug development process, leading to a run. By leveraging models that can identify target disease pathways or predict drug candidates, companies can accelerate preclinical discovery, leading to a breakthrough of product approval. These models could utilize clinical and preclinical data to understand biological mechanisms and molecules of interest, while also developing tool governance and risk mitigation. 

Both the FDA and pharmaceutical companies can shorten the time to market by leveraging AI, leading to faster scientific innovation and better patient care. Insilico has become a leader in this space by using AI to optimize their drug discovery program, requiring only a fraction of the time and money to develop their preclinical idiopathic pulmonary fibrosis product. As this new drug development method gains popularity, it is necessary to consider these novel research methods to remain competitive. 

Thinking Through Your AI Strategy  

With all tool implementations, organizations must consider people and processes in addition to the technology itself. Depending on the use case, these tools might require GxP compliance, necessitating additional validation efforts. Proper governance of the tool may need to be established to standardize tool and data access, protecting preclinical and clinical data. 

To prepare for this institutional investment, an implementation roadmap may be necessary to best consider compliance, governance, and training data. Contact Clarkston’s data strategy experts to begin understanding your artificial intelligence strategy. 

Subscribe to Clarkston's Insights

  • I'm interested in...
  • Clarkston Consulting requests your information to share our research and content with you. You may unsubscribe from these communications at any time.
  • This field is for validation purposes and should be left unchanged.
Tags: Artificial Intelligence