From Hype to Health: Delivering on the Promise of AI in Biopharma


Krishna Cheriath, chief data officer at BMS, outlines the ways in which AI is already transforming the delivery of healthcare, what modern biopharma companies’ AI innovation agendas should include, and why behavioural change is a key tenet in securing meaningful health outcomes with AI.


As Biopharma increases AI investments, to further effective but responsible AI development, leaders need to focus on and invest in a comprehensive and multi-dimensional strategy involving innovation, compliance, standards and processes, and behavioural change management

Artificial intelligence (AI) has moved beyond the hype cycle and is now in the mainstream of efforts to advance healthcare with innovations emerging in research, discovery, support for the patient journey and healthcare delivery. Companies can gain competitive advantage by expanding and accelerating the value proposition of their data, analytic and scientific capabilities through AI augmentation.

The surge of AI in healthcare is evidenced in the increase in venture capital investments in AI healthcare startups. CBinsights reports investments rose to USD 26.9 billion in the first half of 2019. Technology giants like Amazon, Google and Microsoft have made healthcare one of the primary focus areas for AI innovation.

The rapid evolution of AI in healthcare is also evidenced in the significant increase in published scientific research. For example, a recent issue of Nature magazine featured a publication on Synthetic organic chemistry driven by AI seeking to demystify AI for bench chemists to spur future research on how chemical AI will run in the era of digital chemistry. The Mayo Clinic published findings that applying AI to an electrocardiogram results in a simple, affordable early indicator of asymptomatic left ventricular dysfunction.


In the biopharma industry, AI offers exciting opportunities with transformative use cases such as:

  • Accelerating identification and evaluation of promising therapeutic targets. AI image analysis can augment pathologists in rapid and accurate diagnosis.
  • Accelerating and enhancing drug development. AI can accelerate clinical trials, enable robust protocol design and generate insights for precision treatment and improved patient outcomes.
  • Enhancing manufacturing. With smart factories, companies can maximize operational efficiency across assets through AI enabled predictive maintenance.
  • Enhancing supply chain effectiveness. AI can monitor market demand signals, predict shortage risk and recommend optimal inventory levels.
  • Enhancing patient experience. Using AI for patient education – getting answers to frequently asked or searched healthcare questions and for patient adherence
  • Transforming corporate functions like HR, IT, Finance and Procurement to improve productivity and enhance effectiveness. AI is enhancing candidate experience in talent recruitment, automating and augmenting financial processes, detecting cybersecurity threats, alerting the enterprise to supply chain disruptions and more.


The potential for Biopharma AI innovations to deliver meaningful value to patients and the healthcare ecosystem is real. But as Biopharma increases AI investments to further effective but responsible AI development, leaders need to focus on and invest in a comprehensive and multi-dimensional strategy involving innovation, compliance, standards and processes, and behavioural change management.


Companies must have an effective AI innovation agenda that includes:

  • AI use case ideation: Nimble internal processes to accelerate ideation, evaluate ideas for business value and assess ability to execute.
  • Funding mechanism: Fund internal venture capital to finance experimentation and exploration.
  • Data foundation: Trusted, quality, high veracity data is critical. This includes data, technology and talent for building, registering, testing, training and deploying AI models.
  • Stage gates and exit criteria: Clearly define success criteria and stage gates to evaluate the models.
  • Scalability pivot: Define internal processes to pivot successful AI pilots, planning for scalability and subsequent lifecycle management.


Companies must ensure AI operates within legal, regulatory, compliance, privacy, integrity and ethical boundaries. This requires investing to:

  • Establish a clear understanding of legal and regulatory requirements and guidelines, some of which are still evolving. For example, both the U.S. Food and Drug Administration (FDA) and the European Commission are considering frameworks specifically tailored to promote the development of safe and effective medical devices that use advanced AI algorithms.
  • Establish a mechanism to ensure anonymized and identifiable data follows robust consent and privacy expectations. In the European Union, GDPR provides specific requirements on the use of data for “automated individual decision making and profiling.”
  • Establish a data ethics framework to ensure that beyond privacy, data and algorithms consider fairness so algorithmic decisions do not create discriminatory or unjust impacts.


Companies should also focus on AI standards and processes for AI sustainability:

  • Permissibility, explainability, accuracy, auditability and fairness of algorithmic decisions
  • Peer reviews and approvals
  • AI lifecycle management


Finally, companies should also focus on behavioural change management. In a recent JAMA article, Dr Ezekiel J Emmanuel and Dr Robert M Wachter argued that AI advancements in healthcare will not achieve full potential if there is not a meaningful behaviour change. They wrote, “A fundamental challenge facing the US healthcare system is to figure out how to effectively change routines and ensure these changes are embedded in the culture of the system. AI can have a role here, but it will not be simply through better predictions. Instead, the focus needs to be on the ‘effector arm of AI,’ thoughtfully combining the data with behavioural economics and other approaches to support positive behavioural changes.”

This is a powerful point and is particularly relevant in Biopharma. A focus on behavioural change to help leaders and knowledge workers trust and adopt AI augmentation in their workflows and routines is essential to the full realization of AI’s potential in healthcare.

Success in each dimension by itself is not sufficient; rather, prudent attention and investment in all four is essential to ensure the AI efforts in biopharma deliver results and, ultimately, meaningful value to patients. Such a four-dimensional focus will let visionary biopharma companies move AI from hype to health.

Related Content

Latest Report