BACK TO Business ResilienceZINNOV PODCAST | Business Resilience
In March 2023, just 5 months after it launched, ChatGPT cleared the GRE and LSAT exams. Microsoft’s move to integrated ChatGPT into its search engine, Bing, is a bold, strategic move with far-reaching effects on the technology ecosystem. This will not only be a new technology to reckon with, but businesses are now thinking about how to incorporate Generative AI in their workstreams – to save time, optimize costs, and ensure precision. Generative AI has been in the works for years, but only now are we seeing its widespread applicability across use cases and industry verticals.
In this conversation between Matthew Jennings, Vice President of Customer Success, Microsoft, and Rajat Kohli, Partner, Zinnov, the two leaders explore the uses of Generative AI across Agriculture, Healthcare, and how it is poised to make a difference to businesses, teams, and the world at large.
Alongside its rapid growth, Matt also talks about how Microsoft is making a conscious effort to ensure that the data used in its Generative AI applications is inclusive, diverse, and caters to all parts of the society.
Listen to this episode about all things Data and Responsibility in today’s age of Generative AI.
Rajat: Satya Nadella has famously called Artificial Intelligence the defining technology of our times. The experience that AI has been able to create and continues to create our inspiring and these experiences cut across industries and business functions, and AI is changing the very fabric, how businesses are run and how people work. In the recent past, Generative AI has taken the world by storm with newer use cases being explored across the board.
Hello everyone and welcome to an all-new episode of the Zinnov podcast Business Resilience Series. I’m Rajat Kohli, Partner at Zinnov and I’ll be your host for this episode.
As enterprises and common people like us explore all that AI has to offer, we have Matthew Jennings, Vice President of Global Customer Success at Microsoft with us today to unravel the nuances of how quickly AI is evolving and what it means for enterprises. Matt leads a global team that accelerates customer value and sustainable transformation by delivering a connected global experience enduring relationship management and state-of-the-art solutions. Prior to Microsoft, Matt was a Global Vice President at SAP Americas driving global go-to-market initiatives. He has global experience across a variety of industries, segments, and domains including manufacturing, IOT, start-ups, enterprise, digital transformation, and sales and customer success.
A very warm welcome to you, Matt. Great to have you with us today.
Matt: Rajat, thank you for the warm introduction there. I’m very happy to be here.
Rajat: Let’s dive into the world of AI with you Matt, how it’s impacting the different industries, the use cases and hear from you. Generative AI is a rapidly evolving field that has the potential to transform many industries and applications.
However, as with any emerging technology there are several key burning issues that needs to be addressed, such as bias, fairness, ethics, transparency, and IP among others. How are technology giants addressing some of these challenges?
Matt: You think about AI and the impact that it’ll have. It’s really, in our view, one of the defining technologies of our time. And what does that mean for business? But more importantly, how do we use it responsibly? Like how do we put the right practices in place to allow the appropriate guardrails, if you will, to drive the right outcome and how we focus on the most challenging business issues that we face today?
I think one of the important things is that data is a key element for AI to really function appropriately. And the question is really how do we leverage the data to know that one, it’s secure data, it’s valid data, because you need those things to drive the right outcomes. If you think about the balanced presentation that you need for organizations and trying to work it through a model, like it needs to be a balanced model, it needs to be all inclusive of all the aspects of thought. When you think about the impact that it’ll have on people, process, design, culture, the future, there’s really a lot of things that need to be taken into consideration as you look at that.
AI is, if it’s new, it’s like the electricity, you know what I mean? And then data is really the grid that it runs on. So how can that be representative of everybody in all the different thoughts that we have to maximize the ability to make these decisions and be all inclusive. I think the best perspective on it is… we have an AI Officer at Microsoft and Natasha Crampton. The comment there is really with the right guardrails, like this cutting-edge technology can be applied in the appropriate way to really be productive and go on and solve some of our most pressing problems. So really that’s our perspective on how we look at some of this cutting edge technology.
Rajat: Interesting, Matt. I think you touched upon the people and the process as well and during the COVID times everyone was talking about the W3 – Work, Worker, and Workplace. But if you look at now, how do you think ChatGPT AI will impact the workplace and the workforce? Which are the industries that will benefit the most by these technologies and why?
Matt: You know, there’s a lot of chatter on that you think about ChatGPT and the discussion of will this eliminate jobs and what’s the impact on that. I’ll go back to say with any technology that you look at, whether it be the steam engine or the calculator that came, it wasn’t that people forgot how to do those things or were replaced. It was that they used better tools to apply their task and their skill to solve some of our most complex problems. It frees them up to do other tasks. If you think about, especially in my role about the experience, Customer Experience and Customer Success, how do I free myself up from some of those mundane tasks or administrative tasks to go focus on what’s really important to drive outcomes for our customers and our employees and those types of things?
I think if you look at industries like education, there’s going be different ways about how the technology is going to be applied there as opposed to elsewhere. If you look at some of the examples that we have, ChatGPT has already passed the MBA exam, for example, so and it’s already passed Medical Board licensing exams and things like that.
So we probably then need to put those guardrails in place about how do we protect, against making sure we get the right outcome and right balance. If you look at some industries like real estate, you can write those property descriptions and those summaries probably in a much more holistic way through an AI engine like this. Or if you look at marketing or maybe job descriptions and things like that, how do we make sure that they’re more inclusive of everything that needs to be included, as opposed to maybe from a human sense where you might leave something out, not intentionally, but you didn’t have a well-rounded, rooted knowledge of all the data that’s possible to include, if that makes sense.
Rajat: Interesting. Other than the real estate industry, maybe we’ll pick one of the citizen-sensitive industry which is healthcare and get into the future of healthcare. Different technologies have helped improve patient outcomes, reduce costs, and overall the quality of care. And AI was instrumental in combating the recent COVID-19 crisis to a larger extent. Do you think we have become more dependent on AI and other newish technologies, including the Blockchain and Cloud? What are the advancements happening in this space?
Matt: Well, healthcare is one of the most important, probably where we can apply data most appropriately to have the biggest impact.
You think about people’s lives, you think about diagnosis and how do we diagnose and the amount of data that’s there. I think that probably the human body is probably one of the most complex elements or processes that we have to deal with because of all the different things happening together that needs to happen in uniform.
I think the more data we about the different diagnoses, the different clinical trials, the different imaging technologies, those things need to come together. So it really is tangential to not only a diagnosis, but a diagnosis with, and what is the with that you need to include? It could be a clinical trial, it could be the ability to read an image from an x-ray, what are all those elements that come together and all that data that needs to exist? I don’t think any longer you’ll have the orthopaedic being separate from the cardiologist. And what are the relationships between those two specialties? I think we need to bring those together and I think what we’ll look at is probably longer opportunity for life if you will, but also opportunity to rapidly increase maybe the application of, you know, what your symptoms are and how do you relieve those faster with better outcomes. And more opportunity for misdiagnosis and things like that, just based on the data and the technology to help read and evaluate the data.
Rajat: I think our audience is very curious to know, Matt, like what will be the applicability of the ChatGPT in such industries like healthcare.
Matt: I think if you could ask some things, like if we just put down the conditions of what my symptoms might be, maybe that’s a starting point to start somebody in the right direction to say, ‘Hey, I really didn’t think about X, Y, Z, if you will’, by having some recommendations there.
Or maybe as a doctor I specialize in, say cardiology, but I don’t know enough at the moment about oncology. And so maybe ChatGPT can make some recommendations to say, ‘Hey, the three things and the symptoms that are pressing from a cardiologist perspective, there might be some recommendations, you might want to bring some oncology expertise into this.’
So there’s some relationships you may not draw just based on your field of study and based on your field of expertise that when you look at the holistic amount of data there’s other things that can be concluded from that. So I think that’s probably one of the major benefits that we’re going to see.
Rajat: Interesting. AI is again one of the fundamental technologies which is driving the future of mobility, especially with the connected cars. Ecosystem is now talking about the planes, moon, platooning, etc.
What is the role of Big Data, AI, Cloud in enabling Mobility as a Service to a larger. Section of the society in a more efficient and convenient way than the conventional transportation models?
Matt: Yeah, that’s a really good question. I think mobility is probably one of the biggest areas where we’re going to have an impact. You know, if you think about the evolution of mobility, the things were connected. We had a connected car which allowed you to do some other things like sharing… automotive sharing, mobility sharing, whether it be a vehicle, a car, a scooter, whatever that is, like the ride sharing capabilities. But now autonomous.
If you think about connected, the ability to capture data, the ability to share that data so you can do other things and utilize assets that might be idle in a regular sense, like a car in a driveway and then to autonomous… When you think about the ability of an autonomous vehicle and the data that it requires there is a tremendous amount of processing there that needs to happen.
If you look at Microsoft’s relationship with GM Cruise, look at some of the work that we’re doing at Volkswagen about having that cloud platform and how do we develop that together and then you also go back to some of the things I talked about, like there’s other elements of data that you probably need to bring into that.
So it’s what the vehicle is doing, it’s what the driver is doing and producing their data. What are the surrounding environments around you in the data that’s there? Could be weather patterns, it could be construction patterns, it could be other services that the cities are providing, whether it be refuse pickup, could be construction.
But you start to think about all those tangential data elements that need to come into autonomous driving, so that you have a safe and productive journey that’s the most efficient that it can possibly be. So you think about high performance computing, definitely needed. You think about cloud environments, definitely needed. And the opportunity to pull those data elements together to infer good decisions becomes paramount in the ability to produce good autonomous driving results.
Rajat: Though there have been several new use cases, Matt, being explored in driving mobility forward most of these are still in the nascent stage. But they have definitely created many opportunities across the value chain. What are your views on the rise of the connectivity enabling newer supply chains for AVs and the EVs? What are your thoughts?
Matt: If you think about the opportunity of autonomous driving, like if you back up from there, what are the things that need to be in place to do that?
You think about EVs as an example. You truly need to connect. You need the charge. You need to do software updates which is radically different from a traditional, you know, gasoline engine environment. And so how do you pull those different elements together? You also need to look at the ability to develop the supply chain to supply the chip sets that need to go into those vehicles.
You need to look at how do you do connectivity on a reliable, secure, high speed basis. But you also need to look at the impact on the supply chain that that has and on manufacturing, because all those elements have to work really well together in order to be effectively. And more importantly, if you look at autonomous driving, it becomes really about safety.
You want to have a safe journey as to where you’re going, and so you need to make the proper decisions. That’s going to lead into innovation and sensors, innovations in technology, innovations in cloud computing. But you think about that, all those elements need to come together to be good strong decision-making capabilities and how do you drive that? And AI is a primer of that, but also the cloud computing and the processing, the backend high performance computing are all elements to do that.
Rajat: And now like one of the interesting and very close to everyone’s heart is the agriculture industry and considering how critical agriculture is to sustaining life, there’s a significant opportunity to do things more efficiently, intelligently, and sustainably. What are some of the tech innovations that can result into a positive impact?
Matt: Yeah, when you think about agriculture, that’s really interesting because, you know, similar to autonomous driving lives depend on autonomous driving, people being. With agriculture, you do farming and if you have a low yield, that impacts the food supply, if you will. And so one of the things that we need to do is look at all the other patterns that play into that.
And so, if you have a critical storm, a hurricane that comes through a certain environment or a country, it has an impact on the yields. You think about the insurance providers that provide crop insurance. You think about the distributors, the trucks that distribute the crops to the stores and the processing facilities and those types of things.
It has a ripple effect through the environment. So the idea, if you look at some of the investments that Microsoft has made with some of these companies who study data patterns, if you will, can you get really good at the terminating data pattern? Provide better, maybe lower cost crop insurance, maybe higher yields in areas that had low yields, just because you can manage those weather patterns and do the high processing of that data to have a broader impact.
And so, not only are you serving better from a food security standpoint and getting higher yields, but you’re impacting that entire supply chain. And it’s typically in environments and countries where maybe the employment isn’t as high as they would like, but you can start to put some security not only from an agriculture perspective, but in the culture itself in general or in the entire population, because then you’re providing high reliable jobs and those types of things by processing the data in a much functional way.
Rajat: So my next question is all somewhat related to what you said. It generates large amounts of data, agri-tech, which can be used to improve the decision making and increase efficiency. Do you see a huge demand coming up for Cloud and AI applications to manage and churn business insights out of these large data sets. Do you think so?
Matt: Well, I absolutely do. I often say, if you look at a connected environment, it’s really just a way to introduce data traditionally not seen into the enterprise. In order for the enterprise to make the best decision you possibly can, in order for AI to really be as efficient and effective as you possibly can, more data is better. And I often say that sometimes data elements and data sources that traditionally would not be connected to that data environment, if you will, the process needs to be connected in, because there’s relationships that are not known, that are not inherently traditionally known, or maybe not holistically known just based on where we are in our current environment.
You know, I mentioned the oncologist to the cardiologist. Oftentimes there’s probably not a connection there, but you bring in that vast amounts of data and the better decision making you can have by driving these AI models.
Rajat: I think, we touched about, Matt, data. Now data plays a very critical element in all these industries and the applicability of AI as well.
One challenge that you highlighted maybe is the legacy or the data connectivity, but are there any other challenges that the enterprises are facing today with respect to bringing the data together that can be addressed by the technology giants like Microsoft.
Matt: Well, I think there’s a couple of things. We have this idea around security called Zero Trust and you need to make sure that that data is secure. You know, one of the challenges that you have if you’re going to run a strong AI model is that you have to have good data. Once the data becomes unreliable, your whole model becomes unreliable.
So from a security standpoint, you really need to drive the right outcomes to make sure that the data is reliable, is secure, and is available. Then you can start to build the right models to go drive that. The connectivity element you talked about, things need to be connected to be able to transfer that data. You still have that security element from a connection standpoint, but you also need to have very reliable networks and whether it be a satellite network, a cellular network or just a landline network, those have to be secure as well. So I think it’s no longer just about on-premise compute, now it’s about that global compute, global connectivity, global data, global security around the data and those types of things.
Rajat: Interesting. And my last question, as you part of this Customer Success business function and you’re engaging with the customers on a regular basis, what are their key priorities for this?
Matt: Well, the key priorities for this year is we always say, how do we achieve more with less? I mean, can we make less investments and be more efficient and effective about the outcomes that we drive? That seems to be top of mind for a lot of individuals. That does the right things. You look at your data a little bit differently.
I wouldn’t say we’re necessarily applying AI models yet, but we are looking at a lot of digital strategies with our customers about how do they take some of those processes that might be more administrative, as we’ve talked about and make those digital. And then how do they engage with their customers differently.
When you do that differently and you do that via digital, that’s also generating data. And so now you have data on how you’re interacting with customers, what impact that has had, how do you continue to refine that? You know, it certainly is not a destination point to become digital. It’s more of an iterative process because you become digital, more data, become more intelligent, make better decisions, change your processes again. It’s an iterative process to drive efficiency.
Rajat: Interesting. Great, Matt, this has been a wonderful conversation. Thank you for sharing your perspectives with us so candidly. I’m sure our listeners found it as insightful as I did. Thank you once again for taking the time to be here with us.
Matt: Great. It was a pleasure. Anytime.
Watch this episode with Pari Natarajan and Alan Trefler, about digital transformation's evolution in autonomous enterprises. The leaders discuss how decentralized decision-making can increase efficiency, and ensure business growth.
In this episode of the Zinnov podcast, Pari Natarajan and James E Heppelmann explore the effects of Artificial Intelligence on physical design, right from predictive analytics and 3D generative technology to computer vision.