6 Questions with Shahram Ebadollahi, Enko’s Newest Board Member
June 22, 2022
By Jacqueline Heard
We're delighted to welcome another new member to our board: Shahram Ebadollahi is an AI and data science expert who has spent his career applying these technologies to disciplines including drug discovery and healthcare. He previously served as Vice President, Innovation & Chief Science Officer for IBM Watson Health; today, he is the Chief Data and AI Officer at pharmaceutical company Novartis.
Shahram recently shared what intrigues him about AI’s potential in agriculture, why startups have a unique opportunity to integrate AI in their business processes and how they can thoughtfully apply ethics principles when using it. Below is a condensed and edited version of our conversation.
Tell us about how you came to join Enko’s board and what you most look forward to in the role.
Enko approached me in the process of adding independent board members. My background is in applying AI to human healthcare, and I saw many parallels to drug discovery in Enko’s applications of AI to agriculture. Given everything happening on a global scale in agriculture, I see this work as critical and am eager to help guide Enko, especially on the potential of AI in this space.
What are a few of your biggest learnings about the value of big data and AI/ML from your time at IBM Watson Health and your current role as Head of Data Science and AI at Novartis?
Today, my primary focus is on how to infuse AI into the operations of large corporations. At Novartis, I examine different aspects of our pipelines and operations and try to embed AI-based algorithms to elevate the quality of our decision-making.
The massive growth in data and the power of computing infrastructure is what makes this possible across industries. The AI algorithms are at a maturity level that means we can use them efficiently and rapidly, and adjust to the needs of a given organization. For a company in any industry—including agriculture—to make better decisions and develop better products requires paying specific attention to how we can embed AI in every aspect of our work.
As you know, we’re delving more deeply into AI as we incorporate a great deal of new data into our machine learning models. What parallels do you see between early applications and burgeoning uses of AI in the pharma and agriculture industries?
The biggest parallels I see are in the early stage of drug discovery and design. In life sciences, designing molecules for eventual use by patients is all about generating data and interrogating various structures. Then, we apply AI to design something that has the desired capabilities and characteristics. While our endpoint might be different in ag, this early design and discovery process is very similar.
Are there problems in agriculture or food security that you’re particularly motivated to solve?
I see many parallels to the problems I’m trying to solve in life sciences. There, my goal is to prevent someone from getting sick or help them get healthy again. This work in agriculture is a continuation of the same idea. It’s about helping people live a healthy life, and food plays an important role in that.
Given the severity of climate change and how it will affect our food supply, innovations in agriculture are much needed. I’m very interested in the role data and AI can play to find better solutions to feeding people around the world.
How can an organization go from recognizing that it has a great deal of data, to implementing AI and seeing real business impact?
There’s a lot of hype around AI. Many organizations jump on the bandwagon to apply it, but their efforts never see the light of day. I think the solution is critically thinking about how to embed AI into a company’s core operations and decision making. In big organizations, that’s the biggest challenge because it requires changing the way you work. At a large company, I’d say 70 percent of the work is a change of mindset and approach.
Startups are different, because they have the unique opportunity to build everything with AI in mind from the very start so that they’re not playing catchup in a few years. As an independent board member at Enko, part of my role is to guide the company in that direction. It’s much more difficult to add AI later.
How do you think about measuring the value of AI or data science?
In some industries, like finance or consumer goods, this is easy because you can calculate AI’s impact on sales. But in an industry like agriculture, there’s a big buffer between when a company leverages AI to develop a product and when the end user applies it. The question is: How much of the product’s final value can we attribute to AI?
Here’s how I prefer to think about it: if a company uses AI to enhance the accuracy, quality or agility of its decision making, that has value in and of itself. It means the organization will be more efficient and thoughtful about investing time and money, which ultimately creates more value.