Executive Conversations: Generative AI for responsible innovation in pharma commercialization

EVERSANA is a leading provider of commercial services to the life sciences industry. The organization is building generative artificial intelligence (AI) applications to help pharmaceutical organizations drive efficiencies and business value while improving patient outcomes. Scott Synder, chief digital officer at EVERSANA, recently spoke to Ujjwal Ratan, data science and ML leader for healthcare and life sciences at Amazon Web Services (AWS), on the promise and potential of this groundbreaking technology for commercial teams in life sciences, and the guardrails needed to promote responsible innovation in the industry.

This Executive Conversation is part of a series of discussions held with leaders who are pushing the frontiers of the healthcare and life sciences industry with AWS technology.

Ujjwal Ratan: Welcome Scott. To get us started, could you share the core mission of EVERSANA?

Scott Snyder: Our mission is to help life sciences companies manage the launch and commercialization of products, or address specific program and patient needs. We provide integrated commercial services solutions for customers that range from one-product biotech companies to large multinational pharma corporations. We create and execute successful brand strategies across the entire commercial life-cycle from market access, marketing and field services, medical affairs, patient support, and supply chains that deliver significant business impact and better patient outcomes.

UR: That’s amazing. How critical is AI and ML to deliver on this mission?

SS: It’s extremely critical, since every organization in the life sciences industry is considering AI and ML to improve patient experiences, digitize business operations, and make better data-driven decisions. The challenges we solve – from drug pricing, promotion, access, reimbursement, adherence, or product delivery – can be re-imagined with AI to make them more efficient and targeted. So, it’s a huge area of opportunity for us.

UR: You recently published an article with AWS that used the phrase, “pharmatizing AI for the life sciences industry“. What does it mean to “pharmatize” AI?

SS: Pharmatizing means overlaying the unique needs, requirements, and goals of pharma onto the innovation capabilities of generative AI for responsible and ethical innovation. We must take advantage of new tools to do things with transparency, trust, safety, security, and privacy front and center, while innovating at speed to develop and deploy high-impact solutions across the pharmaceutical commercialization value chain. For example, using AI to create relevant personalized content and patient interactions based on their history, while keeping their data secure and preventing potentially harmful recommendations.

UR: Really powerful. What are some use cases you are looking to enable?

SS: There are a number of planned use cases, since our platform touches the entire commercial value chain, from pre-launch to post-launch, for over 650 brands. The initial planned applications include:

• Medical and regulatory review process solutions to help optimize time-consuming, manual compliance operations.
• Field and patient assistance solutions like chatbots, to automate mundane tasks, provide more accurate responses, and improve user experiences.
• Disease and product education content generation and personalization to help life sciences brands improve engagement and education for healthcare providers and patients.

UR: That’s great to hear. How are you collaborating with AWS to bring these generative AI use cases to life?

SS: The collaboration between EVERSANA and AWS focuses on high-impact generative AI applications across pharmaceutical commercialization. We’re using Amazon Bedrock to unlock a range of use cases that require different foundational models, from enterprise search to content generation. We like the approach AWS took with Bedrock, to not only offer a populated set of models from its ecosystem of partners, but also the option to bring your own model. This gives us a flexible environment where we can experiment, with the built-in security and scalability of a high-performance cloud infrastructure, to meet the unique requirements of this industry.

Secondly, we appreciate that AWS shares our values around privacy, transparency, security, and responsible innovation. This allows us to ensure that the right guardrails are in place to benefit patients.

And most importantly, AWS’ industry solutions have a high credibility among the organizations we serve, which is critical. We’re pushing into new frontiers, so trust is everything.

UR: This is a great segue to my next question. Generative AI holds huge potential. But its success depends a lot on an effective data flywheel in the backend. How should organizations design their data systems to take advantage of these technologies?

SS: A solid data strategy and governance plan is non-negotiable. You need a framework to ensure the underlying data is a living asset that gets updated regularly, fuels effective feedback and training loops, and has stewardship. At the same time, outside of typical data and management mechanisms, you’ll need effective foundations to bring structured and unstructured data together—so you’ll need to build new layers on top of the data foundation you already have.

A common misunderstanding is that there is no longer a need for data scientists­. That is incorrect. Organizations will need data scientists to think of data more holistically. And data scientists need up-to-date skills around model tuning, prompt engineering, and overall AI literacy to bring this new vision to  life.

UR: What steps should organizations take to future-proof themselves as they build teams for leveraging generative AI?

SS: Teams will need a pyramid of capabilities that span awareness, proficiency, and mastery.

At a baseline, all team members should have a basic awareness of and literacy around generative AI tools, and how those things impact their workflows. At the proficiency level, they’ll need people who can configure or build on top of these generative AI tools for what I call “little I” innovations. These are small, incremental changes that adapt generative AI tools to solve specific business challenges. Finally, they’ll need some experts, or people who have a deep understanding of generative AI. They know how to train models on a specific corpus of data, understand what good results look like, and can build machine learning operations over time.

It’s also important to have a robust, responsible AI framework that’s shared with all employees so everyone understands the philosophy and guardrails as they innovate.

UR: The space is moving at a pace where what we’re talking about this week could be obsolete the next. How do you separate the hype from the hope?

SS: We need to cut through the noise and triangulate. Generative AI does an extremely good job of predicting the best plausible result based on what a user provides. But we still need human judgment, reasoning, and expertise. AI might get us 80% of the way there as a brainstorming partner, but a critical 20% requires humans in the loop to mitigate risk. Companies that find the optimal balance of human- and AI-powered services will leap ahead in every industry, and pharma is no different.

UR: Before we close, how do you see the space evolving, and what are some promising applications in pharma commercialization that you’re excited about?

SS: There is a lot of potential—everything from customer experiences and omnichannel engagement to software development and operating models can be reimagined with generative AI. However, it’s important to see what’s practical versus what’s still too uncertain.

For example,  accelerating approvals by automating traditionally manual processes like medical and regulatory reviews of marketing content. Another area is patient assistance programs, where generative AI can be used for customizing patient outreach and omnichannel campaigns, offering patients conversational self-service tools, and to personalize programs. On the other side, for healthcare providers, there’s massive potential for improving disease and product education through personalized, on-demand content.

What I’m most excited about is that generative AI breaks down an intimidating layer of data science and opens it up for everyone to use. It lets anyone become a data scientist and interact with data in more intuitive ways to unlock innovation.

UR: Thank you so much for your time today, Scott. I’m excited about the possibilities of generative AI and its transformative potential for the life sciences industry. I look forward to seeing how this space evolves in the future.

Learn more about how EVERSANA and AWS are working together to “pharmatize” Artificial Intelligence across the Life Sciences Industry.

To learn more about how AWS is helping customers innovate across healthcare and life sciences, visit AWS for Healthcare & Life Sciences.