IP Innovators

Trust but Verify: AI as Every Attorney’s Second Pair of Eyes

• DeepIP • Season 1 • Episode 4

Share episode

In this episode of IP Innovators, host Steve Brachmann sits down with Mark Kesslen, Partner and Chair of the IP Group at Lowenstein Sandler, to trace his path from electrical engineering to chief patent counsel on Wall Street—and how that fintech vantage point shapes his practice today. We dig into LabMorgan’s early online payments bets, the ripple effects of State Street → Alice → Recentive, and where AI is already moving the needle in prosecution (IDS automation, office-action drafting, claim-charting) — plus the security and workflow realities firms must get right. 🚀

About the Guest:

Mark Kesslen advises venture-backed startups and iconic tech brands on patents, trademarks, and IP strategy, with deep expertise in fintech and proptech. Former Chief Patent Counsel at JPMorgan Chase.

🔧 Sponsored by DeepIP — your trusted AI patent assistant: https://www.deepip.ai/

🛎️ Subscribe to the channel for more real stories at the intersection of law, tech, and AI. 🎙️

Steve Brachmann:

Hello, my name is Steve Brockman and welcome to IP Innovators, where we profile the careers of patent attorneys and explore how technology and legal practice has advanced with time. This podcast is brought to you by DeepIP, the first trusted AI-powered patent assistant integrated into Microsoft Word. Deepip helps patent professionals draft better patents faster and analyze office actions to craft responses that win. By handling tedious tasks, deep IP frees you to focus on delivering higher quality work and greater value to your clients. Fully customizable, it adapts to your style so it sounds like you only quicker and sharper. Trusted by top law firms and in-house teams across the US and Europe. Today I am joined by Mark Kesslin, partner and chair of the Intellectual Property Group at Lowenstein Sandler. Mark, thanks for joining us today.

Mark Kesslen:

Thanks for having me.

Steve Brachmann:

Absolutely so. Can you give a little bit of your background and how you got into patent law, your journey into the profession?

Mark Kesslen:

So it was my sophomore year as an electrical engineer that I came to the realization that I love technology but didn't want to be an engineer and I had no lawyers in my family, no professionals in my family, and I kind of talked to a number of professors and one of them introduced me to the idea of patent law. So what I did was I finished my electrical engineering degree but at the same time I beefed up my writing skills because I definitely needed to learn how to write. I went on to law school and in my 1L summer I worked at a boutique patent prosecution shop in Boston, massachusetts, and really found that was a great connection for me and that continued on with my career, leading to me to be a patent lawyer at also a boutique in New York City that merged with a larger firm about three or four years in. I then was looking for a needle in the haystack job. After spending time at a firm where I prosecuted patents, I litigated patents, I did the same trademarks copyrights I became really a full service IP lawyer. But I was looking to go in-house for a period of time and see what it was like on the other side of the table, and there's not a lot of chief patent counsel jobs in New York City.

Mark Kesslen:

At the time this was in the late 90s. However, the banks were going through a period of transformation. One of my friends from my engineering school went over to Citibank to become the chief patent counsel. I became the chief patent counsel at Chase that then became JPMorgan Chase, and another member of my engineering class became the chief patent counsel at Morgan Stanley. It was really odd that three people from the same engineering class all had chief patent jobs in New York City at the large banks.

Mark Kesslen:

During this period there was this transformation. There was something at Lab Morgan which was kind of an incubator accelerator that they had set up to really start seeding new fintech companies. That was a tremendous amount of fun when you're kind of there. At the beginning. They had a number of successful exits. I think five of the companies ultimately went public. But then the dot-com boom kind of busted and that kind of shut down and I was looking for my next thing. Obviously, I still had a pretty. I had a great job at JP Morgan but was looking to go back to do the early stage work and found my way at Lowenstein Sandler.

Steve Brachmann:

Are you able, if we can, talk a little bit about your time with Lab Morgan? Can you talk about some of the technologies that were being developed there and if you can be specific about anything that you were able, that jp morgan was able to spit off?

Mark Kesslen:

uh, through lamb morgan sure so, going back to lamb morgan for a second, I think one of the things that was really interesting is you were at a time period where everything was very much brick and mortar. This is the time period of 1999 to 2001. And what? It was? Also a period post State Street decision, which was the case that enabled software to be. We've seen lots of changes in the law since then, but that was kind of the first big case.

Mark Kesslen:

It had to do with the management of funds in a mutual fund, in Lab Morgan. We were innovating online bill pay. We were creating new exchanges for in the bond market, in the foreign exchange business. One of the ones that I remember working on and some filing patents specifically was in enabling a consumer to go to a car dealership and complete and kind of a loan application or a credit application, all online and get some relative immediacy of acceptance or denial of that claim, that company. That was the baseline of a company that was ultimately spun out. And you know it was the platform that many car dealerships used subsequently for years to onboard new customers and get them to go. But there were a number of others at the same time, particularly in the investment stage, that saw that kind of success. Um, candidly, for every one success I'm making up a number. There were 20 or 30 failures.

Mark Kesslen:

But that's the nature of portfolio theory when you're starting new companies and it was quite an exciting time because, when you think about it, you had new payment systems, you had a new bill pay, you were looking at new credit card reward programs, which there was a lot of litigation in the space at the time litigation in the space, believe it or not and it was this realignment of the role technology play in financial services. And if you fast forward to today, I would say one of the two of my busiest parts of my practice as an IP lawyer remained fintech and now property tech, management of real estate and things like that. It continues to see innovation and obviously AI running through that.

Steve Brachmann:

So, taking a focus on fintech now, one of the things that's nice when you've had a career in one particular area of technology is you really get a good sense of how that tech has developed over time. In your view and this can just be your personal opinion are there any areas of fintech that you've seen develop over the past? However long you've been in that sector, they really stand out to you that you find to be maybe not more novel, maybe just more. They've changed our world today in ways we don't exactly realize.

Mark Kesslen:

The answer is yes, however. The answer why I say yes, however, however, is a few moments ago, I mentioned the state street case. Since the state street case, right, right, you've seen a lot of cases and the patent office really not like in the financial services space, right, the argument right or wrong and we can debate that is it's a business method. So, so a lot of the patents that have been filed, whether they were on a financial product, a tax avoidance system or an overall new kind of method for processing money or funding money or making investments, the patent office has taken a very aggressive role against those types of patents. Right, and so you've clearly seen innovation. Right, because there's been a sea change in both how financial products get bought and sold meaning, just think about some of the trading systems right, the quant stuff that goes on the algorithmic-based trading right.

Mark Kesslen:

Those aren't patentable. Yet it has fundamentally changed the way financial services are done right, so that's why I say yes.

Steve Brachmann:

However, and we've talked a bit about State Street here. I do wonder, given that Alice, alice B CLS Bank had at the center of its case this third-party intermediated escrow system and Alice is really what set the foundation of current step one, step two jurisprudence on subject matter eligibility and how that framework and inquiry is handled. It almost puts fintech under the crosshairs more, because the seminal case in abstract idea judicial exceptions to subject matter eligibility is about a fintech innovation.

Mark Kesslen:

Exactly which made it challenging. So if you looked at the financial services companies that were building big portfolios at the time, they had to take a hard look at everything they had on file and it led to the paring back or the cutting of those portfolios.

Steve Brachmann:

Have you found the way that you advise clients over time? If you could speak generally about this, have you found that it's changed? Or how have you found that it's changed, given that pendulum swing of subject matter eligibility in FinTech from State Street through Alice, and what does that advice generally sound like?

Mark Kesslen:

It's a great question. I think it varies in part, but I spend a lot of my time working with early stage companies and the way I describe investors to early stage companies generally and I'm a little harsher or stronger in the fintech space particularly or fintech companies related is investors definitely fall into three buckets. Those investors don't care. If you have a patent portfolio is bucket one. Those in bucket two are they want to check a box. Did you file a patent? Do you think about it? And bucket three they want to do a deep dive. Right, they really care about it.

Mark Kesslen:

What I find with investors in this space most of the investors fall into buckets one or two in the early stage because they're really investing in the idea and the people and can they get success.

Mark Kesslen:

But, as you know, stephen, once you launch something or put it into commercial use, you've got this bar date coming up.

Mark Kesslen:

Right, in the US, you have up to one year if you, you lost the right to file.

Mark Kesslen:

We're a first to file country, not even mention Europe and other countries which have absolute novelty standards, right.

Mark Kesslen:

So the bucket three investors or buyers of companies, right, in my experience, tend to fall in bucket three, right, which is they want to know that, if they're going to write a big check or buy the company for a lot of money, that you've really thought through the patent issue and have a real strategy right. So, as a result, even though in the early days filing may make no sense to your investors, we have to figure out a way to assess if there's something we can put on file. That's kind of foundational to this new idea, and sometimes you can and sometimes you can't, because really it's just taking something that was done I'm oversimplifying in the physical world and using your ordinary computer to implement. I think AI was changing it, except that the recent case with the guy I just forget the name of the case the recent AI case that came down in April, where you're merely putting a new application on a convention at LLM, they said, uh-uh, that's abstract and not permissible, right.

Steve Brachmann:

Is that Resentive? Resentive beef box yes, okay.

Mark Kesslen:

Yes, the resentive case. Thank you for reminding me. So the resentive case we were looking at helping a lot of clients where they had a new application. Those things were going through. We wrote them really robustly. I think a lot of our cases are going to survive because we added a lot of technology to those cases.

Mark Kesslen:

But of our cases are going to survive because we added a lot of technology to those cases, but there's a lot of them that you're going to see fall very quickly as a result of the recentive case. For sure, you've got State Street, you've got Alice, you've got recentive and you've got a lot of cases in between that seem to not that recentive was targeted financial services, but it impacted financial services specifically as well. I know recentive was targeted financial services, but it impacted financial services specifically as well.

Steve Brachmann:

I know Resentive is. It's definitely a machine learning application. It's more about generation of network maps and event schedules, which, yes, obviously has financial transactional, because that's a lot for, like advertisers yeah, and the technology at base there. I had just gone through the petition for rehearing and I can't remember how many cases it cites. But that case, recentive, comes out in April and it's already made its way down into district court. So recentive is citing a bunch of cases where machine learning technologies are being excluded on this new sort of application of Section 101. Right.

Mark Kesslen:

And to me I've always had a problem, just as we're talking here. I've always thought 101 is the wrong test. I've always thought 102, 103, one oh three is probably the better way to tackle it. And I and this goes right back to the days from Alice I think most of your cases would have probably gotten rejected on one oh three grounds, but we always kind of frame it in one oh one and I I find that even the decision in recintive mixed things between 101 and 103. And I think 103 is the much clearer way, in my opinion, to attack those things.

Steve Brachmann:

Yeah, and recentive makes the point that they're collapsing a lot of the other subject matter eligibility inquiry into step one, when really the novelty issue of is something well-known, routine, conventional that's a step two question. So not only looking at the areas of technology that you've prosecuted over time, but as an early career associate attorney and dealing with handling deadlines, anything that because of the progress of technology today we use cloud computing and AI in practice, but 20 years ago we didn't have that. What were the problems you faced because of that?

Mark Kesslen:

I mean my favorite one will always be we had to hit the midnight deadline for the patent office, or a 10 pm cutoff, to get to FedEx. I work in New York City, so things are a little bit later and post offices are open till midnight and FedEx you know, 10 pm.

Mark Kesslen:

Vivid memories of you know, at 1158, hans you know, running over to the Penn Station post office to get applications on file because I had an absolute bar date. I remember another case where I was a second year associate. It was we were responding to a preliminary in a preliminary junction matter in north carolina and I worked with a partner who always did things last minute and I had to get the plea, the, because you didn't have, you couldn't file things electronically.

Mark Kesslen:

Everything was you had to have paper and I remember running down to get to the fedex station again in the penn station area and I missed the deadline by four minutes and they said you can't get it. And I said to the guy but I'm going to lose my job. And they said you snooze, you lose, basically. And I went back into the garage and basically I bribed a FedEx guy with a $20 bill and said you've got to get this into the system or I lose my job. I mean I was being dramatic to make the point, but for $20 back in the early nineties it enabled me to get my brief that the partner I was working with working late to get down to North Carolina for the next day to be filed. So obviously technology with filing at the US district courts is a lot easier. How you file patent applications is a heck of a lot easier. How you file patent applications is a heck of a lot easier. That's the one that was a huge pain point for me as a young associate, for sure.

Steve Brachmann:

Is there an area of technology that has been incorporated into legal practice that you find in your day-to-day tasks is the most impactful, whether that's just email, whether you find yourself using the cloud all the time?

Mark Kesslen:

So I've started to use I have it pinned on my toolbar one of the new AI tools that's used in the legal industry and I find that it can digest an agreement for me. It can look at a patent application for me. I can ask a legal question. It is like having a legal consigliere just sitting up on my screen all day long. I can send it something to. Can you identify all typos and inconsistencies in the documents and we'll come back with a list. I mean it is saving me so much time. I'm having it read lots of different things. I still think we're at a period for this type of technology to use the old adage trust but verify. I like the trust by verify statement because it is a way that you can quickly get a sense of what you're looking at, what you're reading and what you have to tackle in front of you. But at the same time, I use it more as a checklist very frequently and it's been super valuable.

Steve Brachmann:

What particular tasks do you find your firm applying AI solutions to the most?

Mark Kesslen:

your firm applying AI solutions to the most. So one of the things we realized is we manage a lot of large patent portfolios uh, for some iconic Silicon Valley brands and the information disclosure statement process was just killing us right. There are so many related cases. The duty of disclosure is big. The fees that the clients are willing to pay are kind of on a fixed fee basis, very low, and it was very much ripe for a technology tool. We spent a lot of time with the vendor to be able to upload show, family relationship and what was taking an admin. I'm making up a time period, let's call it one to four hours. The system now does it in 15 minutes. 20 minutes yes, it's a trust, but verify. But it has enabled us to really speed it. Not only speed the process, keep the backlog down, but look at our staffing models as well as a way of managing that and putting people to their best and highest use, which is how we're reallocating resources.

Steve Brachmann:

Great. I think we had, in our pre-recording call, talked a little bit about how Lowenstein Sandler is using some in-house solutions that they've developed and that they're also using some vendors. Without going into specifics on who's who here, can you at least give a sense of what has the firm been able or found themselves able to develop in-house, versus where are you looking outside?

Mark Kesslen:

I mean, I think, quite frankly, in the builder buy scenario, we have generally, as a firm, taken the buy route. I think, when you look at the speed in which technology is evolving is super quickly, while we're a super quick and while we're a big firm you know 400 lawyers you know we're not big enough to compete in on scale at all what we are doing is we're test driving a lot of different products right now. Um, specifically, I'll refer to products in the patent, in the, in the patent universe, we are looking at products that are helping us draft patent applications. We're looking at products that take a claim and can convert them into the drawings for the patent application. We're looking at tools, and not only looking, but using and testing. I should say tools that are looking at an office action and not just creating a template, but using and testing. I should say Tools that are looking at an office action and not just creating a template but actually putting together a response. It might not be in the voice, it might provide too much information, but it is conquering a blank page right.

Mark Kesslen:

We're working on tools that help with figuring as you're changing numbers and part names, ensuring consistency and identifying and when you're looking up terms, can pull in a description of what the particular widget is that needs to connect. There's a lot of great stuff there. We're also kicking the tires, most recently on an AI product that will you need to invalid. You get a cease and desist letter to create a quick claim chart based on prior art. It looks at the art that's out there, both patent art and non-patent and NPLs, non-patent literature and putting together a claim chart with cited references in a really quick and efficient way to enable you to advise the client, like at least directionally where you should go Right. So there's a ton happening. There's a lot of players. I think our team that's looking at this probably and maybe it's a slight exaggeration can speak to a new vendor every day, but maybe it's not every day. It's a couple of weeks every week and I'm also seeing this bleed into, since we're talking IP also in the trademark space as well.

Steve Brachmann:

Can you so do?

Mark Kesslen:

they overlap a lot as far, far as they're different tools. But right right, but? But? But think about you're doing a trademark search, right? I? I'm, we're in zoom, right? I need to search if workplace is available for use and registration for a new, you know whatever right it can create. There's tools out there that are doing those trademark searches for the term workplace, for you know whatever it is in this type of article.

Steve Brachmann:

Right, right. And then just to go back to something you said earlier that I think kind of describes so much of the difficulty of getting started on, you know, briefing or whatever you're trying to do as an attorney conquering the blank page and knowing where to start and I feel, like AI, one of the best benefits of that is it doesn't have to think about where to start, it just goes and does it and sometimes getting that kickstart makes the product things just roll along easier.

Mark Kesslen:

A hundred percent right. So so it's a. So I think. For an associate who's writing patent applications, I think the two best parts of it right now are the conquering the blank page and the final output. And the conquering the blank page is exactly how you described it, steven. The second part, what I mean by the output right, there's still a lot of thinking that goes in and making sure it reads right, the claims are crafted right, but you should never be handing in to the partner who's reviewing it a document that's got typos, inconsistencies of terms, mislabeling, right, because these tools can fix that stuff right. It's the brain-powered part in the middle that's still being worked on, but those two end points in the drafting stages are there to save a lot of time.

Steve Brachmann:

Yeah, yeah. What kind of issues have you had to overcome in implementing new technologies, and specifically AI, into practice?

Mark Kesslen:

I think one of the ones that we're seeing with the early stage clients is confidentiality, security, protection of PII, right, and you've got a lot of issues there. Nothing, really. We're not even kicking tires if it doesn't satisfy our security review, and we've seen a number of early stage companies that haven't met our security standards. I know there's law firms out there that aren't using AI at all because they're afraid of security at all, because they're afraid of security. Most of these tools are in an Azure or AWS private cloud instance. Those are world-class secured places. I'm not worried about that. I am worried about other parts of the security, like what have they done? How have they coded? How have they secured it? Where does it sit? Who did it right?

Mark Kesslen:

I think we are a firm right now, because of our size again, are not on the bleeding edge, but very much on the leading edge. We've got top-down support and bottom-up support and are really making an effort to like. Every discussion is when you're, when an associate gets an assignment, is what can AI do to help you? Right, you need to ask yourself that question because we're trying to change hearts and minds and any I think any any attorney who's not playing with this right now, um is missing out on huge opportunities and and not to use the phrase from earlier, if you snooze, you lose, but I think that's the reality of the AI space in our industry today.

Steve Brachmann:

Yeah, no, that is a note well taken. Can you put into numbers any of the productivity gains you've seen at Lowenstein Sandler through adoption of AI tools?

Mark Kesslen:

I think the best one we've seen was in our information disclosure statement. We really saw material change there. I think the next place we'll see material change is the creating of kind of the template response to an office action right when the office action comes in the prior arts there you know the claims and to kind of take that and put together the form response. I think that that step of kind of conquering the blank page will be done incredibly quickly. Will it be right Different issue right, but that will see changes. Will it be right Different issue, right, but that will see changes. The application stage is still a TBD but I think it's real and I think it's a year ago. It wasn't close, in my opinion, wasn't really ready. I think every three or four months now it's leapfrogging in some exponential.

Steve Brachmann:

Right, the pace of change is so quick in the field.

Mark Kesslen:

And I think candidly, when you think I mean there's lots of different practices of law. In my opinion, patent prosecution is, you know, on the short list where it makes sense, yeah, to be one of the earliest adopters for most change, right? Not only because of the subject matter but, candidly, also because of the professionals who are doing it. They know the technology, they understand the technology, they're not afraid of the technology, and I think that combination makes it a great one-two punch.

Steve Brachmann:

Yeah, and then you have all the scientific literature that you just don't have in other areas of the law and to analyze that, the tools to be able to get quick insights from hundreds or thousands of pages in a pharmaceutical specification, I agree 100%. Yeah, Pros or cons of the current state of AI technology. Are there things that you really love or are there things that you'd like to see developers change?

Mark Kesslen:

It's hard for me to answer your question directly because, you know, to some extent I'm a neophyte in the space, learning to use it in real time. With that said, I think compute power is impacting some of it, right, the ability for it to compare two documents quickly. It wants you to break it into stages, right, and that is solely based on the compute power of the various applications that are out there. I do think that will be overcome. Of the various applications that are out there, I do think that will be overcome. So that's the one that I see, but I think that's just one of timing more than anything.

Steve Brachmann:

Yeah, yeah, well, great, we're getting to the end of our time here together. Mark, I always like to take a minute or two at the end of each episode, having talked about the law and technology, to get a sense of what life beyond the law is like. I have a real belief that you kind of need a hobby in order to make all the dense reading go down. So is there anything for you, mark, that gives you your zen, that helps you get to another day of practice?

Mark Kesslen:

Absolutely so. I'm an outdoor road warrior on my bike, so I'm a cyclist. On weekends it's harder for me to do it. I'm also a Peloton junkie, so since the beginning of days I'm an early riser and work out very early in the morning. I think I do that for two reasons One, because it's good for you and I kind of like sweating. But the other thing I find that that's some of my best time in the morning to think. Um, even if I'm working out hard, um, I come up with some of my best ideas. Siri's great on my wrist, because if I have an idea I can tell Siri to make a note of it, cause if not, um my, it might go out of my brain 15 minutes later. Um, but I do some great thinking and I, I'm a cyclist, but I tend to ride by myself.

Mark Kesslen:

It's kind of like my time to recharge, and I'm a big, I love being outdoors, so the combination just works really well for me. Is there a favorite trail you've ever biked? I did part of the Ironman trail at Hawaii I did, which was spectacularly gorgeous. It was probably one of my favorite rides. Or my other one was kind of through the farms of Provence with a dear friend of mine. So those are probably my two favorite rides, without a doubt.

Steve Brachmann:

I'll have to put Provence on the list. Haven't been to Europe, but France sounds great, it's a beautiful country. Mark, thank you so much for your time here today. This was a great conversation.

Mark Kesslen:

My pleasure, stephen, nice to meet you, nice talking to you.

Steve Brachmann:

Absolutely. I'm Steve Brockman for IP Innovators. Happy patenting.

People on this episode

Head Shot

DeepIP

Host
Head Shot

Steve Brachmann

Host