Powering Progress Episode 9 – Protecting the Advantage

Aubrey Bishai [00:00:09]

Welcome back to Powering Progress, a podcast series exploring the infrastructure, energy, and market dynamics shaping the next generation of data centers and AI development. I’m your host, Aubrey Bishai, Chief Innovation Officer at Vinson & Elkins.

As AI and data center development continue to accelerate, much of the conversation has focused on power, infrastructure, financing, and scale. But behind every major technological shift is another critical issue that organizations increasingly have to navigate.

Innovation governance is the framework we use at Vinson & Elkins to help companies identify, protect, manage, and ultimately maximize the value of their intellectual property, data, technology, and proprietary systems, including protecting against claims from other rights holders. And in today’s environment, that challenge is becoming more complex, and more important than ever.

Modern data centers are no longer just physical infrastructure assets. They sit at the intersection of software, hardware, AI models, cooling technologies, chip architecture, operational processes, and proprietary engineering.

As companies race to innovate, collaborate, and scale, the risks surrounding intellectual property ownership protection, licensing, and competitive advantage are evolving just as quickly.

Today, I’m joined by Hilary Preston, a partner at Vinson & Elkins and Co-Head of our Intellectual Property practice, to discuss how companies should think about intellectual property risks and opportunities in the rapidly expanding AI and data center ecosystem.

We’ll explore why innovation governance matters, where organizations may be most exposed, and how collaboration and partnerships can create new complexities around ownership and confidentiality, and why protecting innovation is increasingly becoming a core business and infrastructure strategy, not just a legal consideration. Hilary, welcome to the podcast.

Hilary Preston [00:02:47]

Thank you. Glad to be here.

Aubrey Bishai [00:02:49]

To start broadly, when you talk about innovation governance, what does that actually mean in the context of AI infrastructure and modern data center development?

Hilary Preston [00:03:00]

Yeah, great question. For us, innovation governance really means — it’s an umbrella term that we use to encourage companies to look at the entire picture of their innovation landscape, and the components of that are essentially all driven towards being forward-looking.

So, taking stock of what IP you own, what IP you are licensing, what IP you might be co-developing with other participants in the project, how you are protected on your contractual rights — ideally, if there’s time, which is an important part of this challenge that we ought to talk about — but when there’s time, looking at the competitive landscape to see what other projects are doing in terms of their IP protection, and essentially trying to gauge where the biggest risk areas are that you ought to pay the most attention to.

Aubrey Bishai [00:03:54]

Data centers today involve a complex mix of proprietary technologies, from cooling systems and chip architecture to AI models, software, and operational processes. Where are you seeing the biggest intellectual property risks emerge? And how important are patents, trade secrets, and licensing strategies becoming in this evolving ecosystem?

Hilary Preston [00:04:22]

That’s a very good question. There is, from a financial and an operational risk perspective, no question that the biggest area to watch out for is unforeseen patent litigation disputes. I would say there are probably four technology areas where we are seeing a lot of enforcement activity, which to us means those are the areas that you ought to focus on as you’re trying to reduce that risk on the front end of a project.

So those tech spaces are cooling and thermal management, power management and energy, networking and AI semiconductors, and then sort of more broadly, the software and the security and the data management that is inherent in an operational data center. But from a risk perspective, we focus a lot on how patent litigation risk is either resolved before it becomes unmanageable, or ideally avoided by way of licensing. And the reason we do that is that is where we see the largest money judgments, and perhaps most importantly for developers and for operators, that is the area where there is the most acute risk that you could end up with an injunction that stops the project, which is obviously a result everybody wants to avoid.

Aubrey Bishai [00:05:42]

Many of these projects include multiple stakeholders working together — developers, hyperscalers, cloud providers, equipment manufacturers, energy companies, and technology vendors. How does collaboration increase the complexity around ownership and protection of intellectual property?

Hilary Preston [00:06:04]

To be blunt, the challenge is that a lot of times these projects move so quickly, people do not have clarity around who owns what, and therefore who’s going to stand behind what.

So what do I mean by that? If you have one participant in the project development providing the semiconductors, one participant in the project development providing the cooling systems, someone else providing the power management — when you have that combination of products and services that are all coming together in potentially a high-profile project, what the patent holders and IP holders see from the outside is an opportunity to assert rights across all of those things altogether.

So what does that mean from a complexity perspective? It means that if the deal was put together and the project came together at a pace in which there wasn’t clear allocation of rights and obligations — who stands behind the technology that’s providing each of those components — you end up in a challenging situation of trying to figure that out on the back end. And when you’ve got the amount of capital that we’re all seeing being deployed or committed, even just by the hyperscalers in this space, you can be certain that it’s drawing the attention of rights holders who are looking to see whether they have some enforcement activity that they can do now, or that they would prefer to do after the project actually gets built.

Aubrey Bishai [00:07:35]

This is certainly complex. So looking ahead, do you think intellectual property and innovation governance will become a larger boardroom and business strategy issue as AI infrastructure investment continues to accelerate?

Hilary Preston [00:07:52]

It will be forced to be one. And what I mean by that is, either it is a situation that project developers and all of the firms that are involved in that effort can look at and spend time on, on the front end, and try and do their best to reduce the risk of future disputes and to protect their own IP that they develop along the way. Meaning, are they getting patent protection on new innovations? Are they taking the steps that they need to take to protect certain categories of information as trade secrets? Are they being deliberate about that process on the front end — which, depending on the level of investment that’s related to that aspect of the project development, they’re either going to focus on it on the front end, or unfortunately, they may well have to deal with it five to ten years from now when disputes arise.

You know, and one of the areas that we are watching very closely is, how is the patent landscape changing in this space? Perhaps not surprisingly, it is changing very quickly in those four core areas I talked about: cooling technologies, power management, networking and AI semiconductors, and sort of that overall software, security, and data management category of IP.

In recent history, there has been an absolute exponential spike in the number of patent applications that are being filed in this space. And something like 75 percent — I’ve even seen higher than that — 75 percent of those applications are being prosecuted by entities seeking that IP protection outside the U.S. Predominantly, those firms prosecuting those IP rights are Chinese companies. And so what happens oftentimes with that scenario, looking down the road several years for the patents to issue — once they issue, we see as a common fact pattern those Chinese innovators, or at least the ones that prosecuted the patent applications, transfer those rights to something called an NPE, or a non-practicing entity, or more disparagingly, a patent troll, that will then enforce them as the project actually comes to fruition.

Aubrey Bishai [00:10:13]

Thank you so much for your insights today, Hilary. I appreciate it.

Hilary Preston [00:10:17]

It’s my pleasure.

Aubrey Bishai [00:10:20]

And thank you all for listening to Powering Progress. To learn more about the issues shaping AI and data center infrastructure, visit velaw.com.