The demand for data centers driven by AI and cloud migration have strained the energy infrastructure. As power consumption exceeds availability, the legal and business challenges of securing power, building out infrastructure and allocating risk have become key issues for the industry.
In this episode of the "Real Assets Roundup," Jones Day partners Brian Sedlak, Paul Jones, Jeff Schlegel and Melissa Vandewater discuss the energy and utility issues facing the data center industry and where future solutions may be found.
Podcast: Play in new window | Download
Dave Dalton:
If you're a regular JONES DAY TALKS® listener, you know Tim Curry and Taylor Stevens. Both are Jones Day partners who advise clients on venture capital, growth equity financings, and mergers and acquisitions, and both are frequent and popular contributors to JONES DAY TALKS®.
Today, Taylor and Tim will talk with partner, Ka-on Lee. She's the co-leader of Jones Day's intellectual property transactions team. Taylor and Tim will ask her about due diligence as it relates to IP and venture deals. Stay here. This is interesting and timely. I'm Dave Dalton. You're listening to JONES DAY TALKS®.
Tim Curry has represented startup, emerging growth and public technology companies and public offerings, venture capital and growth equity financings and mergers and acquisitions for more than 30 years. He represents a broad range of technology companies at all growth stages. Taylor Stevens has, for more than 25 years, advised companies in the areas of venture capital transactions, M&A and a and capital markets transactions.
His practice is focused on the representation of emerging growth and established technology and life sciences companies. And as mentioned a minute ago, Ka-on Li is a co-leader of Jones Day's IP transactions team. She works closely with clients to successfully structure and execute IP technology and commercial transactions to achieve high-stakes business objectives. Thank you all for being here. Lots to talk about today. So, Taylor, kick us off.
Taylor Stevens:
Thanks, Dave. Appreciate it. Great to be back here again with Tim Curry. We've done a number of these podcasts together, focusing on various topics across VC investments and emerging companies' practice. We have a special podcast because we have our colleague and partner who specializes in IP, Ka-on Li, with us today, so we're really excited to have her join us with the podcast.
Tim Curry:
Thanks, Taylor. Yeah, it is exciting that way. I think our first guest on the pod, so this is awesome. It's fitting because we all work so closely with Ka-on and her team that doing the first one with her as a guest, it makes perfect sense.
Taylor Stevens:
So today what we're going to really focus on is the due diligence side of the venture financings and specifically the IP issues that can come up from time to time. So from the corporate lens, for people like Tim and I, when we're doing these financings, I would say diligence issues that come up in the deal fall into one of two buckets.
The first really pertain to capitalization or contract issues, maybe their employee issues, founders have left, or there could be some general litigation. That could be something that's known where we need to look into a particular matter and get comfortable with it or something that comes up during the course of the financing.
But suffice it to say, everything in bucket number one are things that we know how to navigate and they usually aren't that problematic. But then that brings us to a whole different category, which is the second bucket, and this is titled IP issues.
Tim Curry:
So, Ka-on, what are some of the key IP issues and what are the things you're looking for when you're representing an investor in a venture capital transaction?
Ka-on Li:
When we're representing an investor, the first thing I want to do, my first goal is to really assess whether the company's IP assets align with the value proposition that the VC is investing in and whether there are any issues that could undermine the value of the target company, whether there are issues that we can help fix after the deal closes.
And so at the outset, understanding the technology that the target company is based on the foundational products that they're trying to make money on essentially, and the value of that product from the VC's perspective. And then assessing what type of IP normally covers that product, whether it's patents, copyrights, trade secrets, for example. So understanding that it's a software product, how that software product is deployed will also help me understand whether the customer agreements are done properly.
Has the target company licensed to the customer more than what they need? Do they even need to license any IP to the customer? If it's just a SaaS products for example, or if it's a life sciences company, obviously patents are going to be paramount. So that's really the first goal is to understand the value proposition of the target company and how IP is relevant to the company. And then we also understanding the target company's business strategy.
We evaluate whether the IP that the company holds supports the company's commercial roadmap. For example, figuring out whether patents are filed in the right markets, whether licensing or collaboration agreements could restrict future growth or exits. And then also ownership. We verify that the company actually owns the IP that it claims it owns, making sure that there are proper assignments from founders, employees, contractors.
There are no lingering rights with universities, prior employers, collaborators, and of course have they developed their core IP using AI tools. That is something that we definitely dig into more so in the last few years. And we identify issues such as the reliance on open source software without proper compliance with the relevant open source license terms, whether jointly developed IP grants the company adequate rights to commercialize the product. And of course, if there are pending disputes that could affect evaluation as well.
Tim Curry:
The use of AI tools to develop software is a relatively new phenomenon and a really important one in this day and age. So what are you looking for in that and what are some of the pitfalls that you've seen where you've had to dig in a little deeper?
Ka-on Li:
If you ask a company whether they're using AI tools and they say, no, that's already a red flag, because they're obviously not aware that their employees are using AI tools. Every company's employees or third party contractors are likely using AI tools in some form. So that's the first question we ask is how they manage and control their developers and R&D personnel's use of AI tools. Do they have a good handle of what AI tools they're using? And this is important for a number of reasons.
The first reason is under at least US IP laws, if you're generating a particular technology or content or software, for example, using an AI tool without adequate transformation of that output, that output cannot be protected by relevant IP laws. That is what the Copyright office has found under the Copyright act in the US. Same with the Patent office as well.
When it comes to patents, if the company has used AI tools to develop their core IP, that is potentially an issue because they could potentially not have any ownership rights over that output. The second issue is infringement. Some AI tool providers will not guarantee or warrant that the outputs don't infringe any third party's IP rights. And so if the company uses those outputs and takes those outputs into their commercial product, there is some risk there as well.
And last but not least, at least from an IP perspective, is the inputting of the company's trade secrets into an AI tool. And whether the act of doing so may mean that those trade secrets are no longer considered trade secrets because the AI tool provider is potentially allowed to disclose those inputs in their outputs, for example, or to train their AI models. Not all AI tool providers are permitted to do that, especially if you have an enterprise subscription to a particular AI tool. The terms might prohibit the AI tool provider from doing that.
That's not to say that any use of an AI tool by a target company is problematic, but understanding how they've used those AI tools, how they've managed the use of their AI tools, and to the extent that they haven't done a very good job in controlling use of the AI tools or having visibility over the usage of AI tools, that's something that can be fixed going forward. So after the investments putting in place a proper AI policy to mitigate those risks.
Taylor Stevens:
Great. That's super helpful. Thank you, Ka-on. So certainly, there's a lot there for VCs to focus on as they're considering IP due diligence. If we switch gears slightly and think about it from the company's lens, how should companies best prepare for the IP diligence? And look, are more patents equal to more value?
Ka-on Li:
The company should be ready to talk about what their most valuable IP assets are and how they protect those assets. Be ready to understand IP terminology, be ready to understand the gaps in your IP protection and to explain them.
So I would suggest talking to an experienced IP deal lawyer who has worked on a ton of VC deals and M&A deals so they can educate you and help you prepare for those tough discussions and diligence questions you'll be getting as part of the due diligence process is definitely worthwhile.
To answer your question, no, ultimately, it's not about counting patents, especially if the target company is just in the software industry, it's not even expected that they have any patents. It really depends on the underlying technology. When it comes to patents, the question is whether they file a patent which cover the company's core technology, and is that the type of technology that is normally patented versus keeping it a trade secret?
So being prepared to explain how your patents cover your core technologies or to explain why you don't have any patents. And the explanation could be as simple as no, we've discussed with patent council that the best way to protect that technology is by keeping a trade secret. And we protect our trade secrets by implementing adequate, confidential and security practices that align with standard industry practices. Could be as simple as that.
Things like trademarks, you might have registrations on your core brands, but you might not. There might be reasons why you don't have a registration on a trademark, and that doesn't mean that you don't have protection in your trademark. It just means you don't have a registration, but you should have a strategy to protect that unregistered trademark. And then the low-hanging fruit, I would say, is making sure you have all of the paperwork done.
So all of the IP assignment agreements with the founders employees, your third-party contractors, to the extent you've in-licensed some foundational IP from a university or another third party, making sure those agreements are very robust and you have a narrative regarding any issues that the VCs council might identify in those agreements.
And then to the extent you know that you don't have an open-source policy or an AI policy to have a narrative about that, maybe you can say that to the extent it's accurate that you're in the process of implementing an open source policy and an AI policy. That would be a reasonable response depending on how mature your company is.
Essentially, investors don't expect that your IP strategy and implementation of that strategy is going to be perfect when you are such an early stage company. But they do want to know that you've thought about the issues and that you understand the importance that IP has in the value of your company, and you have invested some reasonable resources to protect that IP as well.
Tim Curry:
One particular thread that seems to be found is the lack of getting enforceable proprietary information and invention agreements from overseas employees, so employees or consultants that are outside the United States. And the point on that is that every country and even some states within countries have different laws on what makes a valid assignment from an employee.
And it's not sufficient to use your California PIA form in India or in Singapore or in France or in Germany. There's going to need to be tailoring of those forms, and to the extent that you out there that are founding companies can take care of that upfront, it will do away with quite a bit of questioning and concerns that'll be raised by the PCs.
So another area that I know that we both spend a lot of time in when we're doing IP diligence, and this is more for companies that are commercializing their products, so they've already got some customers and some revenue, but are there commercial contracts? So what are you looking for in a SaaS agreement or a license agreement or some other type of commercialization agreement that companies would tend to have?
Ka-on Li:
Sure. So the customer, a agreements need to be tailored to the product being commercialized and how that product is being deployed. That's in the software space at least anyway. And so understanding how the software is deployed, whether it's a SaaS product, whether it's deployed on an on-premise basis or maybe it's both because then you need to have the right paper to reflect how the software is deployed.
If the software is simply provided on a SaaS basis, then you probably don't need a software license, for example. So that would be the wrong form to use for that particular product. I also want to understand if there is AI being deployed with the product, if there are any AI enabled features or functionalities where the data being ingested by the software could potentially be used to train the target company's AI models and to further develop its products and services, that data is potentially so valuable.
Now, it's not just the revenue that the target company is obtaining from the relevant customer for the software subscription or license, but it's potentially also the data that they're harvesting through their customer relationships. And of course, if they're harvesting personal data, they need to comply with relevant privacy laws and the relevant jurisdictions from which they're collecting that personal data.
But what I'm seeing a lot is these target companies leaving a lot on the table by not receiving adequate rights from their customers to use aggregated and de-identified data to further develop and improve their products. I'm also going to be looking at whether the company's commercial model aligns with what's in the agreement. If they're just charging their customers on a yearly or monthly subscription basis, then I don't want to really be seeing terms relating to user seats, user locations, continental licenses, things like that.
If I'm seeing a mismatch between the commercial parameters around the agreement and what's in the agreement, that's also potentially a flag for me that the target company hasn't really thought about the best way to commercialize their product. And at the end of the day, these customer agreements are assets. The target company is deriving value from these contracts, and if the contracts raise risks for the target company, then the value is very much diluted.
So for example, if they're agreeing to unlimited liability with respect to their breaches, then the relevant customer agreement could actually raise more risk than potential value.
Tim Curry:
One of the things that occurs to me is the value in working with your outside law firm's technology transactions partner like a con in order to get these contracts right, extract the right value, and not increase your risk. And it's not an area to just go on AI and have them generate a contract. It's one to invest part of your legal spend in because it is an asset of the company, like Ka-on say.
Ka-on Li:
Absolutely. And I would say that even if you are contracting with customers who are not going to be using your form agreement and want to use their paper, for example, if you're contracting with a hospital or a financial institution, a large bank, and they're insisting on using their paper, having your form agreement in your back pocket as a baseline for your positions is really worthwhile.
And having a consistent approach across all of your customer agreements offering similar service levels, so you're not constantly having to monitor different service level compliance provisions in different agreements, for example, not only is good contracting practice, but will also help your bottom line by ensuring you don't need to pile on a bunch of resources to manage compliance with different customer agreements with different compliance standards.
Taylor Stevens:
Yeah, terrific. All really helpful. So if we take a step back and think about due diligence from a high level, as the corporate team members, we always have to ask ourselves, what's the appropriate level of due diligence that we're going to do in connection with the financing?
And we may go back and tie out the capitalization, or we may do targeted diligence from the last priced rounds that they did, or we may even navigate what I refer to as reactive diligence where we go in and actually let the company deliver the disclosure schedule to us first, and then we look through it to determine if there's anything that we need to react to.
But we always need to navigate this balance and I think determine the right effort, and that's going to be a function of our client's goals and our client's sensitivities. But how do you balance the light touch for due diligence for early stage investments and deeper dives for later rounds?
Ka-on Li:
So it's about judging the target company against other companies in a similar industry at the same stage of development. We're not going to compare a one-year-old startup company with a huge conglomerate of Fortune 500. The diligence and the results of the diligence do not need to be perfect. In fact, even a large company doesn't execute an IP strategy perfectly.
So working with a law firm, both from a VC's perspective and the target company's perspective, who understands the posture that a typical company in this industry at this stage of their life cycle, what they normally and should have in place goes a long way. And also engaging a lawyer who understands the industry as well. So if you have a lawyer who's asking about FTO investigations and counting patents when the target company is just a software company, of course that might be relevant for certain software companies.
But for example, if it's a company that is leveraging third-party LLMs to develop an AI agent, for example, they're probably going to have less patents. And so making sure that you are literate in that space and can explain to the target company's council your IP strategy goes a long way.
Taylor Stevens:
So, Ka-on, it's really helpful to understand how we would approach it in the efforts there because it's really interesting when you think about due diligence in the context of the financing and how that may differ than an M&A deal, because we rely on the reps and warranties that are given to us in connection with the financing, but they generally serve a diligence purpose.
It's unlikely that a VC is going to go sue a portfolio company over a breach of the rep. So the diligence effort becomes all that much more important in contrast to the M&A deals in which we rely on the reps and warranties. And lo and behold, if there's a breach, guess what? We have indemnification, recourse and all these other things.
So I think differentiating the role of diligence and the role of reps and warranties in the financing context and differentiating that from the M&A framework, I think is really important. And it just highlights the emphasis on doing good due diligence in the venture context.
Ka-on Li:
Yeah, ultimately, the difference between diligencing an early stage company in the VC context versus the M&A context is as council for the VC, you understand that the target company is a work in progress, but you also use that to your advantage. There's a lot of work that can continue to be done if you do identify issues, you still have time to fix them, especially early stage companies that have founders and developers who are still with the company.
It's not too late to have them sign proper IP assignment agreements in favor of the company if they're still there. And even if those employees have left, it's not too late to, and those employees didn't sign proper IP assignment agreements, it's not too late to identify what code those former employees or contractors developed and potentially replace that code without having to redo the entire product.
Tim Curry:
And if instead it ends up that the former employees or consultants were in marketing or some other thing that's less core to the company, then oftentimes venture investors will be okay with the fact that you didn't get the IP assignment from that person because it's not core to their technology and their operating.
Ka-on Li:
Exactly.
Tim Curry:
So, Ka-on, one of the things that you and I have seen a lot over the last five or six years is there's really no such thing as a US venture deal anymore. The company in its operations are still oftentimes based in Silicon Valley, based in San Francisco, based in New York City, somewhere in the United States.
But nearly every venture-backed company has some sort of operations overseas. And that relates into the PIIA issue that we discussed earlier. But what other issues can arise if a company's either developing their software or has other operations outside the US and particularly in China?
Ka-on Li:
R&D in the US is very expensive compared to the rest of the world. So it's very common to see a company establish operations outside of the US, whether it's the Czech Republic, China. Back in the day, it was Russia and even Canada. We're seeing a lot of software development now being outsourced to Canada or target companies are establishing subsidiaries in Canada, especially in the gaming industry, for example.
In both software in AI and also life sciences, R&D in China is a lot cheaper. There will be more scrutiny of course, if the company's technology involves particularly regulated, increasingly regulated areas such as semiconductors, which would be impacted by the CHIPS act, AI of course, pharmaceuticals, biotech and data. Getting IP and data from China into the US, there are potentially hurdles, especially for example, like genetic data.
There are additional steps that the company must take to transfer that type of data into the US. Sometimes we see inadequate intercompany agreements, so between the Chinese subsidiary and the US entity such that the US entity doesn't have adequate rights to the IP being generated outside of the US. And so the IP actually stays in China and the US entity isn't able to derive value from that IP.
So making sure you have proper intercompany agreements in place is very important, and to make sure those agreements are reviewed from a tax perspective as well, because there are likely going to be transfer tax issues involved in those arrangements. And when we talk about companies with non-US operations, for example, China. So first, we experienced that in Russia.
When sanctions were imposed on Russia and target companies had to cease their operations in Russia, transitioning their development team from Russia quickly to Ukraine for example, or to some other eastern European country. And that happened very quickly for the companies that were doing that in Russia. We might be seeing the same thing in China depending on what the Trump administration does. I recently came back from a trip in China where a lot of Chinese clients we were speaking to were very concerned about the geopolitical issues.
And our Chinese clients were very concerned about what would happen when it comes to not only what the Trump administration is going to do to restrict investments and operations of a US company in China, for example, restrictions on the manufacturing of goods in China. But they were also concerned about what the Chinese government would do in response to what our administration would be doing as well.
And they're starting to think about creative structures they could implement to potentially mitigate the issues arising from those geopolitical issues. So again, engaging legal counsel who is abreast of those geopolitical issues and the likely outcome of various executive orders and bills that come down the pike is very important.
Tim Curry:
Taylor, who knew 35 years ago when you and I started doing these deals that all of a sudden these little venture deals would be steeped in geopolitical ramifications and complications? That is times have changed.
Taylor Stevens:
That is so true. It used to be the emerging company startup and the investors were on the same street in Palo Alto.
Tim Curry:
That's right. That's right.
Dave Dalton:
Okay. Tim, Taylor, Ka-on, thanks. That was a lot to cover. I sense there's more information and another podcast coming. Stay tuned. For contact information for any of today's contributors, visit jonesday.com.
And while you're there, check out our insights page for more podcasts, videos, publications, newsletters, blogs, and other important and timely content. Be sure to subscribe to JONES DAY TALKS® so you don't miss any future programs. JONES DAY TALKS® was produced, again, by Tom Condolas. As always, we thank you for listening. I'm Dave Dalton. We'll talk to you next time.
Speaker 5:
Thank you for listening to JONES DAY TALKS®. Comments heard on JONES DAY TALKS® should not be construed as legal advice regarding any specific facts or circumstances. The opinions expressed on JONES DAY TALKS® are those of lawyers appearing on the program and do not necessarily reflect those of the firm. For more information, please visit jonesday.com.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
