Enterprise AI: From Strategic Vision to Real-World Impact

What’s holding enterprises back from turning AI ambition into real-world impact?

At The Six Five Summit: AI Unleashed 2025, Nirankush "Kush" Panchbhai, SVP of Product Management for  ServiceNow's AI Platform, breaks down exactly how to close the gap between strategy and execution. He joins host Daniel Newman for an Enterprise Apps Spotlight with an important conversation on accelerating AI adoption and intelligent system design.

Key takeaways include:

🔹Accelerating AI Adoption & Intelligent Design: Explore the rapid acceleration of AI adoption within enterprises and the critical shift towards designing truly intelligent systems that drive meaningful change.

🔹Balancing Innovation with AI Governance: Understand the complex challenges of governing and scaling AI systems, highlighting the delicate yet crucial balance between fostering innovation and ensuring robust, responsible governance.

🔹Orchestrating Trust & Scale: Learn about the strategic introduction of an "AI control tower" as an innovative approach to orchestrating governance and ensuring trust across the entire enterprise AI landscape.

🔹From Experimentation to Real-World AI Execution: Gain invaluable insights drawn from real-world enterprise experiences, providing a clear roadmap for moving beyond AI experimentation to successful, impactful execution at scale.

‍

Learn more at ServiceNow.

‍

Watch the full video at Six Five Media, and be sure to subscribe to our YouTube channel, so you never miss an episode.

‍

Or listen to the audio here:

Daniel Newman: Hi everyone. Welcome to the Six Five Summit: AI Unleashed. I'm joined today by Kush Panchbhai, SVP of Product Management for ServiceNow's AI platform and this Enterprise App Spotlight Session, we're going to be talking about building strategic vision to real world impact with your AI. Kush, thanks so much for joining. Welcome to the summit, first time. Glad to have you here.

‍

Nirankush Panchbhai: Thank you, Daniel. Thanks for the opportunity. Looking forward to the discussion today.

‍

Daniel Newman: Absolutely. Look, you don't have big shoes to fill. Your CEO, Bill McDermott opened our entire event last year and now we brought you here to keep this conversation going. Last year we were talking about AI, we were talking about how it's going to shift and change the future, but gosh, even in a one-year time, Kush, I can't even tell you how much innovation, disruption, and change has taken place. Even the things we talked about one year ago versus now, it's an incredible time, exponential time. So I want to hear from you, get a lot more of that.

But one of the things that I've really enjoyed in following the journey of ServiceNow is just not only how the company has really changed and disrupted and evolved itself, but seeing the customers getting value. You like to talk a lot about the future of what the software industrial complex looks like, you talk about reducing chair swivels, new ways to bring systems together and make work more productive. So we know organizations are moving towards agentic platforms. As you're watching these companies and these enterprises move this direction, what are some of the common challenges that you're seeing them face in terms of creating this coherent AI agent strategy? And how do you recommend, what do you think they should do to balance innovation while keeping the operation on the rails and minimizing complexity?

‍

Nirankush Panchbhai: So that's an amazing question. I think one of the common themes which I'm observing is there is an organizational tension happening inside the enterprise. Let me explain it to you. So what happens is you have product teams, engineers, designers, PMs, who want to go 10X, they want to innovate, they want to embrace this amazing new technology. At the same time, you have another side of the organization. It could be the compliance team, it could be the legal team, it could be the privacy team. They want to make sure that each and every step they're taking is measured. And when these two sides are colliding, that enterprise is kind of stalling. And one of the things which we encourage customers and we solve at the ground level in our platform is embedding governance in AI platform. And that's why our governance is not bolt-on, on top of it. In each and every layer, we are adding that governance and we have built tools for it.

‍

And it's not just governance, which is giving you visibility. It's workflows embedded in it. So risk and compliance workflows for which we are an industry leader is embedded in the products. At Knowledge this year we launched AI Control Tower. Now this is a single pane of glass in which you can see your entire AI footprint from AI agents to AI skills to AI models. You can govern them, you can see the performance of them, you can see the value it's creating so that both sides of the organizations are running in the same direction for the business strategy. We truly believe that governance is an accelerator rather than a break in the system.

‍

Daniel Newman: Yeah, I tend to agree because the big difference, and we've seen how quickly consumer AI is proliferated and it's basically because most of what we're using has come from open internet data. So the compliance risk was different. This data was already out there, it was already available. And so we know that in the early instantiations of trying to do AI in the enterprise, there was all the, okay, when we start commingling data, do you remember even some of the early LLM use cases where people were putting proprietary data in and it was causing all these problems because now you're feeding something that's constantly learning something maybe it wasn't supposed to know.

‍

Nirankush Panchbhai: Yes.

‍

Daniel Newman: Then now once it knows it, it's like how does that ... So when you're ServiceNow and you're looking at companies, they're saying, "We're feeding critical HR data. We're feeding critical CRM data. We're feeding critical infrastructure systems data about our company." That company that is using your tools has to know that the data is safe. We hear about it whether it's through compliance and governance, through sovereignty. These are different angles and of course trust layers and securities, these are big things in the enterprise, probably has a lot to do why we can't go faster.

‍

Now, I want to talk a little bit about hype here with you too Kush, because we've had this kind of theory of AI and we hear AI is going to ... Everything from it's going to take over the world, it's going to take our jobs, but also it's going to create amazing growth, productivity, efficiency. But we're also sort of hearing that not everything that is theoretically happening is actually being deployed or even being deployed at scale. So how do you think about going from that kind of we want AI to make us 10 times more productive to going to an enterprise and saying here's a practical, scalable approach to start getting outcomes in a timely manner because how fast things are going but in a realistic capacity?

‍

Nirankush Panchbhai: So you used an amazing word; outcome.

‍

Daniel Newman: Yeah.

‍

Nirankush Panchbhai: I think one of the most important things enterprise needs to do is align their business strategy to the AI outcomes. That's one of the things which is not happening naturally. You would see enterprises are looking at how many models they're deploying, how many skills they're deploying. I guess the real question is what value that model brought to your business? What value does skill got your business? So that we move from experimentation to aligning everyone to the business strategy and tackling those outcomes with AI. And I think that's when the real magic happens because you are measuring each and every AI investment, you're figuring out are you going in the right direction? You are governing that AI and seeing the value of that AI. And then you are course correcting it accordingly.

‍

When you are aligning at the top-level business strategy, you're solving it east to west, each and every department is coming in and figuring out, for this business strategy, how am I aligning, how am I pitching in to make sure that this AI is an accelerator and helps us move forward compared to the competition? And I'll give you an amazing example actually. So we have a customer called AstraZeneca. Everyone knows them. AstraZeneca was spending 30 minutes per person on procurement. Now think about the amount of hours gone into procurement rather than researching and building life-saving drugs. So when AstraZeneca standardized on ServiceNow AI platform and automated those workflows, they saved 30,000 hours per year. Now that's real measurable impact because that is all the time gone into researching new diseases, coming up with new drugs, and essentially aligning to the mission of AstraZeneca and creating outcomes which are measurable.

‍

Daniel Newman: Yeah, and I like that you bring a real-world example, companies in fields like healthcare kind of hit on both the things we've talked about in terms of the real-world scale challenges Kush, but they also face the real governance challenges. They deal with a lot of very sensitive data. So they're trying to scale and move these workflows very, very quickly, but they also have a lot of sensitivities, whether it's trial data or customer data. So you've got to really balance all that. Talk to me a little bit about how enterprises should think about building a system of intelligence. Now, that's a little bit of a different word. We've often heard system of record. In the AI era, system intelligence, it needs to be able to scale AI, but it has to really embed what we just talked about, trust and transparency into every layer or else it's just not going to work in this particular era.

‍

Nirankush Panchbhai: So 100% spot on. I think scaling AI is super important. And if you're not doing it right, it's like building a skyscraper on sand. It may look amazing, but as soon as you do a little bit of scrutiny, that skyscraper falls down. And that's why we believe in providing governance and trust in each and every layer. When you can trust anything, your adoption increases with that. If a developer is producing something and you can trust it, you would adopt that features. If regulators can govern that, they will be more comfortable releasing it out there. So that's why with AI Control Tower, it's not just a system of record like you called out. It's a system of intelligence because we are embedding workflows into it, compliance workflows, legal workflows, security workflows, risk workflows, and they are not just something which you treat as a department.

‍

We are taking a east-west-north-south approach because that's how you're going to be able to scale AI now. If you are just going to think about AI as a point solution and not a platform, then you are not going to be able to do complete business transformation with this amazing technology. So when we use AI Control Tower internally, we see how much impact it's creating even for ServiceNow. At ServiceNow, we have a program called Now on Now where we drink our own champagne. So we have seen that our self-service deflection has gone 14% up. In just 120 days when we deployed agentic AI internally, we saw $10 million of savings. That's like having 50 employees' throughput saved with AI, which is super, super amazing. So we see the results internally, we see the results with our customers here. One external example which I'll give you is Bell Canada used ServiceNow AI platform for their customer service and in one year they deflected 3 million customer calls, which is huge. And they did that with compliance embedded in it, so they ran fast with compliance baked in.

‍

Daniel Newman: Yeah, no, I think that's really important, and I like your analogy because I use the east-west analogy a lot too. We've talked a good bit about north-south with the governance and such, they kind have to be built into the systems. In this era we love talking about kind of fast-paced infrastructure. So today you and I have kind of talked about all the things that you should ... Take a breath and a caution and think about as you're kind of building the stack.

But then the second part of it is as you build the stack and then you start to take it across systems, because yes, we are seeing companies like ServiceNow with what you're building, the Control Tower and the now platform, they can sort of aggregate. In the agent era, companies probably are not going to want to have agents running on hundreds of different softwares. What they're going to want to do is build agents centrally, deploy them. And then like I said, some things become databases, some are databases with logic, others is literally just going to be like middlewares and connectors, and all the things that really need to happen to connect all the software in a real enterprise estate.

‍

So you've got to do that and then you've got to govern it. And then the data has to be really well-thought-out, how it moves, because you are responsible and in different regions of the world, data leaks, it's a big risk not just to your customers but to you as well. How are you sort of approaching this challenge of governing, the fact that you're going to increasingly be pulling data ... You're doing Raptor, but I mean eventually everything's going to become the data and ideally if you succeed at ServiceNow, ServiceNow agents are going to scale and span across all the productivity and applications and databases and it is going to become this single pane of glass that we talk about. How do you approach that challenge and what can you share from your current experience about what it's going to take to get that done?

‍

Nirankush Panchbhai: So another amazing question. I think one of the things which you will see from the ServiceNow AI platform is we bring three key ingredients on the platform: AI plus data plus workflows. And that's the key for business transformation because the fuel for AI is data. So like you said, we need to have data connectivity and with Raptor and workflow data fabric, we get that for our customers. On top of that, we build an amazing AI layer which you can govern and it's an open system where you can pick and choose which LLM model, which infrastructure you want to get at. And then we bring in 20 years of our experience in doing workflows. That's how we move work from east, west, north, south. And when you have these three ingredients on that platform, this platform gives you everything you need to do that rewrite you were talking about because that's how you are building new outcomes, you're solving new problems. You're not only solving for operational efficiency, but you are thinking about how do I grow my top line with AI now so that I am growing from both sides of that equation?

‍

And especially with the workflows embedded into the platform, customers are able to run super fast. I gave example of our GRC workflows. Anytime you want to run a risk and compliance workflows, you're able to do that and you are able to do that in a scalable way. Today we have an EU act for AI, but we want to make sure that this platform is scaling to any new act which comes and that's how you do it with these workflows. A new act comes in, you plug it in, and that's where the machinery is churning behind the scenes to make sure that each and every regulation, each and everything the customer cares for is maintained on the platform.

‍

Daniel Newman: Yeah, I think there's a lot of challenges there and I appreciate you trying to simplify. The next couple of years Kush because you're going to be very interesting as companies kind of try to tie all these threads together. And that's why I said I think as much as we want enterprise to move at that sort of speed of sound, it does take the right partners, it does take the right technologies. I think there is going to be some need for some removal of some of the abstractions of how many layers and enterprise has of software and of tools to be able to go as fast as we're going to need to go in the AI era. But it's really compelling and of course all the things that AI creates challenges, it can also be used to help us with all this stuff. Kush, I want to thank you so much for joining me here at this year's Six Five Summit. It's been great chatting to you and we'll have to do it again soon. Really enjoying following the ServiceNow journey.

‍

Nirankush Panchbhai: Likewise. Thank you Daniel.

‍

Daniel Newman: And thank you everyone out there for joining us for this Enterprise Apps Spotlight at the Six Five Summit talking about AI, a lot about governance there, compliance, really important stuff. And while some of that stuff is more ... It feels like it's the nitty-gritty, we are never going to see enterprise scale if we don't consider all of those important details. Stay connected with us on social and explore all the conversations here at sixfivemedia.com/summit. More insights coming up next.

‍

Disclaimer: The Six Five Summit is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we ask that you do not treat us as such.

‍