The Next Autonomous Vehicle M&A Wave Starts Below the Line

Executive Summary

In this episode of The Road to Autonomy podcast, Grayson Brulte sat down with Hugh Nguyen, Partner, Automotive Technology & Mobility, KPMG where they discussed the resurgence of the autonomous vehicle sector and the complex web of partnerships currently driving the industry forward.

Key The Road to Autonomy Episode Questions Answered

Key The Road to Autonomy Topics & Timestamps

[00:00] Autonomy is Back: KPMG on the Industry’s Resurgence 

The autonomous vehicle industry is “back in vogue,” with KPMG seeing a resurgence in interest from automotive, mobility, and technology clients. After releasing their initial “Islands of Autonomy” whitepaper in 2019, KPMG is now helping clients navigate a new wave of direct investments and strategic inquiries as the ecosystem matures.

[03:59] The Strategy Divergence: Robotaxis vs. Personally Owned Vehicles

Immense capital is required to pursue a vertically robotaxi integrated strategy, forcing most developers to choose between robotaxi networks or personally owned vehicles. Hugh notes that for personally owned autonomy to succeed, manufacturers must find the “magic level of investment” that creates a “good enough” self-driving capability at a mass-marketable price point.

[06:50] The Partnership Ecosystem: Dating before you Marry 

Partnerships can be described as “dating before you marry,” allowing companies to test relationships before committing to stronger joint ventures or M&A. However, this rapid “speed dating” is creating unexpected competitive dynamics, where long-standing ecosystem players, such as rental car companies, suddenly find themselves competing with new fleet operators.

[08:58] M&A vs. Partnerships: Understanding the Lack of “Teeth” in Current Agreements

Unlike joint ventures or M&A deals which have strict legal “teeth” and asset definitions, current partnership agreements are often less binding and lack punitive measures for failure. Consequently, the success of these alliances relies entirely on the willingness of operational teams to collaborate rather than legal compulsion.

[16:00] IP Protection: The Paradox of Collaboration and Competition 

A major challenge in these partnerships is the “paradox of collaboration,” where companies must share enough information to work together without revealing proprietary “secret sauce”. Hugh identifies that the most sensitive IP is often not the software code, but the “below the line” operational knowledge, such as safety management standards and fleet operating procedures.

[29:56] “Below the Line” Operations: Why Fleet Ops are the Hidden Scaler 

“Below the line” operations including fleet management, cleaning, and logistics, are just as significant a scaling factor as vehicle supply or technology. As robotaxi deployments expand, the ability to manage these complex, multi-layered networks involving various subcontractors will determine success or failure.

[40:20] The “Precision Strike” Approach to Future Talent Acquisition 

Rather than massive corporate acquisitions, the future of talent strategy will likely involve “precision strikes” to acquire specific individuals or small teams with niche expertise. Companies are specifically targeting talent capable of generating high-quality synthetic datasets and identifying unique edge cases to accelerate development.

[48:46] 12-24 Month Outlook: Consolidation and the Survival of the Fittest 

Looking ahead, the industry expects continued “speed dating” through partnerships for the next 18 to 24 months, though failed partnerships will likely fade quietly without public announcements. Ultimately, the market will see consolidation where the “winners” will be defined by their capitalization and ability to shape policy, will acquire the capabilities of those who cannot sustain their operations.

Full Episode Transcript

Grayson Brulte: Autonomy is back in vogue. It’s cool. It’s not new kids on the block, it’s autonomy’s back on the block. And lo and behold, your firm, wonderful firm, KPMG. I’ve got the hat. I’m not that good of a, a sports player, but I got the hats. How are you thinking about autonomous vehicles?

Hugh Nguyen: That’s right. I’m not that great of a golfer either. Just to set the records straight. Well, you know, it’s interesting. Um, we have a nearly eight decade legacy of serving, um, the automotive ecosystem. Um, in 2019, so several years ago was when we actually put our first, uh, thought piece or white paper out called Islands of Autonomy that was talking about how we viewed, um, autonomous vehicles. Right. Would would. Proliferate Flash forward to today. Most of our customers, particularly, um, you know, large OE globally and, uh, you know, suppliers up and down the Tier Ridge, um, are all either asking questions about autonomous vehicles or are making direct investments in autonomous vehicles or in the ecosystem, or are figuring out, you know, what to do with those investments if they haven’t performed as well. So we are asking ourselves. Much more about how best to serve that ecosystem, how to be knowledgeable about it, and what our automotive clients care about. The other part of it is our mobility clients, our technology clients, our ride share clients are also asking the same questions, so we are taking it seriously because all of our customers are taking it quite seriously. We have to be able to answer their questions.

Grayson Brulte: Your customers have to be able to take it seriously. Elon got the vote. 75% of shareholders overwhelmingly voted for his compensation package. FSD continues to get better. We wrote in it the other day, I have it at home. How much of the, the awareness, the we want to get back, and this is being driven by the, the, you wanna call fomo, the fear missing out because FSD just continues to accelerate in a really positive way.

Hugh Nguyen: Well it, it’s interesting when we look at the metrics by which, uh, the Tesla board have set to guarantee that incentivization package, there’s a minimum vehicles delivered, but there’s also a minimum FST subscriptions. Target like 10 million subscriptions or about, I think eight x of, you know, where it sits today. Um, it’s interesting because as a consumer of that technology, uh, you know, I, I think it’s essential, right? I think it works very well and personally find myself using it fairly often. I live in a geography here in the San Francisco Bay area where, you know, a lot of my neighbors think the same thing, but there’s parts of the country where the, the penetration for that OEM is not as strong. Right. Um, and, uh, you know, folk live in, um, different environments where, um, you know, you maybe can’t necessarily guarantee the performance of it all the time. So, you know, the way I think about it, um, you know, it’s, it’s them, it’s that OE and others that are contemplating how best to invest to get that ADAS capability, um, up to speed, whether they do it themselves, whether they license it out and rely on someone else to do it. Um, it’s a race, right? To get to that capability. Uh, you’re seeing, um, oes that have had different varying strategies in, um, an av whether it’s personally owned or robo taxii, um, change directions, right, and grow and make serious investment. So I think it’s very, very interesting, um, and personally feel as if, uh, there’s going to be quite a, quite a bit of advancement development capability, particularly for the OE that are really invested in it.

Grayson Brulte: It is past a DAS. Yeah. A DAS was, sorry, two years ago. Mm-hmm. Do you see a divergent in the market where if you have your, your Waymo, your Tesla’s an outlier? Mm-hmm. Yep. That are just purely focused on roboto taxes, and then you have your traditional OEMs that are focused on two things. One, contract manufacturing and two personally owned vehicles. ’cause we got the announcement from GM 2028. The Cadillac escalate IQ level three. The key here is eyes off. Hands off. Yep. That’s a big step. And in the press photo, if you noticed it, they had a picture of Netflix in there.

Hugh Nguyen: Yeah.

Grayson Brulte: Yeah. So is it Netflix and Chill in the car? Is that where the oes want to go?

Hugh Nguyen: Well, you know, it’s interesting just thinking about the fragmentation between those that’ll pursue a completely ver vertically integrated strategy versus those who will have to pursue a different, more fragmented, perhaps, partner strategy. Just the level of investment, uh, and capitalization needed. To, to pursue the whole stack, right from the hardware to the self-driving, um, sensory to the data, right, and to the compute needed to do something with that data. Just, um, those are almost insurmountable, insurmountable obstacles, right? For any given individual, you know, companies. So the majority of the players in the space will have to pursue, um, a strategy to achieve autonomy, autonomy. You know, very, very differently. Um, we know, we’ve talked at length about mm-hmm. Um, you know, robotaxis versus personally owned, and the question for personally owned becomes, um, at what point does the r and d needed to create a good, good enough, right. You know, self-driving capability. At a price point that’s sensible and mass marketable. At what point, what, what is that point? Like what is that magic level of investment and what can we do to invest just that much, uh, to enable that capability for, for personally owned? Um, you know, as for, you know, robo Taxii, um, you know, very interesting, lots of exciting partnerships being announced. Um, I feel as if, you know, over the last two weeks, just the, the, the, the headlines and the news and the names and. Everything that’s being announced from market entry, right? And expansion to, um, technology and ride share partnerships and manufacturing partnerships. Um, I think you’ll continue to see, uh, a large spate of partnership announcements and activity, uh, in the coming year or so, which may potentially evolve into as many tech, as many tech, tech cycles do, winners, losers, and potential consolidation down the line.

Grayson Brulte: The, the, the partnership. Ecosystem is on overdrive. It’s basically these partners are, are chugging monster beverages and, and, and they’re going, they’re going full steam ahead. But there’s gonna be, there’s a glut of partnerships. Yeah. And we saw this going back to, let’s go back to, uh, 2019. 2020. Yep. Okay. They imploded. Mm-hmm. When do we get to a point where there’s too many partnerships? Yeah.

Hugh Nguyen: Well, you know, I think it’s, it’s funny ’cause I think it might’ve been you actually that used sort of the. You gotta date before you marry analogy. And that’s what these partnerships are, right? They proceed, um, stronger tie up joint ventures m and a, although like the rash of announcements in the past two weeks, kind of makes it seem like speed dating as opposed to just regular dating, which is, which is, um, which is, um, kind of crazy. But I think, you know, with, with these partnerships, and I think it’s really important to note that, um, you know, whether it’s a network partnership, right? The leverage of, uh, one hardware provider of another’s network. Or the use of one’s hardware. Um, it’s not just that you have, uh, partnerships forming at such a rapid rate. As a result of these partnerships, you’re finding that there are ecosystem players, companies. That had no idea they were competing with others or all of a sudden start to compete with each other. Um, you know, imagine being a, um, well-established, uh, decades old rental car company, right through a virtu by virtue of your partnership with a robot taxi provider, you’re all of a sudden competing with like a pure play fleet operator, right? That wasn’t in the playbook, you know, three, five years ago, and now it’s happening in real time and real speed. Um, often people are finding out as announcements are being made. So I know the question is, you know, how do we figure out whether a partnership’s real? But I, I see the pace of that accelerating faster, you know, over the next six to 12 months before we actually start to see, um, some of the partnerships. Uh, maybe the lack of success of some of those partnerships not be advertised as loudly. Because the thing with these partnerships is, um, you know, it’s, it’s not as if, if the partnership. Doesn’t do as well commercially or doesn’t achieve its aims. You’re not gonna read a press release about it, right?

Grayson Brulte: No.

Hugh Nguyen: Neither partner or partners is gonna come forward with a statement saying, Hey, didn’t work out irreconcilable differences. It kinda just fades into the background, right? And typically it goes the way of, um, you know, uh, acquisitions of capabilities, people leaving, that sort of thing. So I think that’s what we can expect in terms of, um, the proliferation of partnerships going forward.

Grayson Brulte: But what happens when partners become competitors? And we’re seeing that emerge not just in the big one, but we’re seeing that start to emerge with, with several companies now. Mm-hmm.

Hugh Nguyen: What happens then? Well, um, I’ll, I’ll give the perspective from m and MA, an m and a practitioner. I’m an m and a practitioner, um, by trade. I think the key difference to remember between partnership and like pure play m and a, whether it’s an outright acquisition or like a joint venture or some sort of agreement, there’s not a ton of teeth in those partnership. Um, agreements, right? So for example, like if we were um, dealing with two companies entering a joint venture, typically in that joint venture agreement you have to spell out the assets, you’re contributing the percentages. You have to spell out breakage terms, right? There’s a lot you have to spell out ’cause it’s an actual trading contribution of assets to new, to a new entity in, in a partnership agreement. Um, they take many, many shapes and forms. But you don’t necessarily have to spell out things that are that binding, right? The terms aren’t necessarily, um, you know, don’t, don’t, don’t need to be as binding as they might be, uh, with the finality of like a purchase or a sale agreement in an in an m and a. But because of that, right? Because of that, um. The actual success of that partnership, whether it works out or not, because it’s not necessarily punitive to either, for it to not work out ’cause it’s not being spelled out in a, in agreement with teeth just comes down to how much each partner is actually willing to invest at the operational level, you know, to make it happen. And you know, sometimes if you see skeptical reactions from the street or companies to partnerships, um, it tends to be. Hey, you know, how much is each partner actually going to contribute? Or what can they actually bring to the partnership? Is there a clear capability gap that one has that the other has that actually makes it fit? Um, you know, those are the key questions to evaluate whether a partnership is gonna be successful. And as to the point about, um, you know, competition dynamic, I said earlier when, um, the partnerships don’t go as planned, um, you won’t see a pronouncement. Right. Um, and, uh, you know, uh, when, when partnerships don’t go as planned, um, you know, it’s, it’s egg on the face of both, right? So, you know, I think you’ll have to read between the headlines and between the announcements made, um, to actually see whether partnerships live or die. You’ll see it in the form of job descriptions that get posted. You’ll see it in the form of, you know, people moving from competitor to competitor, partner to partner. So you gotta look at it that way.

Grayson Brulte: I have, I have the KPMG hat on today, but I have the AU hat on. That’s right. But I do my inspections and, and reading these things, but So you have the partnerships. Some come, some ghosts, some last, some get divorced, some, some stay married forever. Why is there this rush, rush, rush, rush, rush. Mm-hmm. To put out a press release when you clearly spelled out. You don’t have this, this due diligence document of. He said. She said, because that’s what it comes to. Yeah. Why is there such a rush to push out these press releases? Why not bake the relationship, test it date, and then go public with the press release if it all does work out when, as you said, there’s a clear agreement that could be used in the event of a separation?

Hugh Nguyen: Well, I, I, I think, um. You know, with the proliferation of announcements in that space, uh, I won’t speculate, but you know, there’s a, there’s a headline pop, there’s an effect, right? To a partnership being announced. Um, in the automotive space, you have open a open innovation partnerships. We know we suppliers, startups, and those typically are not broadcasted outwardly. It’s actually just mutual agreement to collaborate on a pilot or a product or something. Right. And to see where it actually goes, that’s very different. That’s, you know, that’s, that’s very different materially than a. Partnership announcement at the same time, you know, it takes work to make a partnership announcement. Typically, when you make a partnership announcement, there’s a volume commitment, there are dates thrown out there, right? There are metrics and measurables that are, um, understood and valued by the investor community to actually, um, you know, uh, have the, the, the, the halo effect of, of a, of a partnership announcement. You know, to how you’re valued. So I, you know, I don’t wanna undermine, um, the work that corporate development, investor relations and product teams right. Have to make and do in order to reach a conclusion to announce a partnership. But I think it tends to boil down to the involvement of like the product teams and the ops folks behind it to understand where the gaps in capabilities are and to like, you know, actually put some substance to that announcement.

Grayson Brulte: If you look at, so you have the, the, the top log goals that just make a decision. But then you have the product team, and you have the information team, you have the policy teams, the legal teams. What happens if some of those lower level teams Yeah. Decide, Hey, I don’t like this person anymore. Yeah. I mean, what, what do they have to replace them? What happens in that circumstance?

Hugh Nguyen: Well, you know, it’s, it’s, it’s difficult because in, in, in a nascent, you know, EV ecosystem, even individuals sort of. Maybe further down the chain, you know, more in the field, more operationally focused as opposed to product decision making or, you know, strategic has knowledge to Right. Um, that, that, that can be difficult to replace. It’s not like there are. You know, a million companies out there each with major fleet operations expertise. Right. It’s still concentrated in the hands and brains of a, a few leaders, and I mean, gray, you know this right? Like the industry is, people kind of know each other on a first name basis. Yeah. Right. Industry is still, you know, very, very small. The ecosystem still has a small town feel. Right. Um, I think what it comes down to is this, uh, one trend that I’ve noticed among, um, uh, among savvy partnership. Formation experts and m and a practitioners is the corporate development teams, right? The, the corporate development teams that are typically tasked with, um, determining, evaluating and executing m and a strategy are all of a sudden becoming very savvy. Product engineering ops, legal, uh, professionals. Um, you know, it used to be the typical pathway was. Banking or big four into corp dev and then you, you drive deals that’s still very much as a pipeline, but now you have, you know, engineering, right. You know, product folk, um, you know, who have built and developed things right? And deployed things, moving into corporate development, um, and taking that route. And I think that that savviness at that level, at the deal making level is, you know, essential to understanding whether a partnership is good from a business development move only or good from an actual value creation perspective.

Grayson Brulte: You hit the nail on the head. Small town. Yeah. There are very few of us that have been in this industry 15, 16 years and we all know each other. Yeah. We friends, our families, our friends, we get together, go on vacation together. So it, it is very small. So a lot of this industry is personal, but I read a blog post recently, uh, A 16 Z Jason Nor has put out on Substack. They are seeing an a significant increase in NDAs and they’re saying that is the first sign towards a brewing m and a activity. Yeah. Are you seeing the same thing that A 16 Z is seeing?

Hugh Nguyen: Well, um, well, I can’t speak to the count of NDAs. I’m asked to review or sign or, you know, observe, clientele, sign. Um, what I can say is this, uh, I did mention earlier that like a, a partnership maybe in terms of the rigor of the agreement can, doesn’t necessarily have to be as, as, as rigorous as, um, like a joint venture or a pure play deal, but. But, um, savvy clientele are applying a little more rigor and focus on how those partnership agreements are actually structured to make sure that they’re not exposing, um, IP to make sure that there’s a way to recover IP or information in the event that a partnership, um, doesn’t end with its objectives met. So I wouldn’t say that there’s a degree of. Adversarial understanding being built into the partnership construct right now. But I actually think that that’s a good thing, right? I think that, um, you know, in this ecosystem, in this small town field where no one’s got curtains, right? Yeah. Um, you know, uh, you can inadvertently, um, reveal a lot of sensitive, proprietary, um, information, uh, to your partners. Without necessarily knowing it. And, and it’s not just, you know, um, the self-driving stack, the, the, the, the engineering behind it, the software, what we traditionally think of as ip, it’s the processes, right? It’s how fleet operations runs, it’s how depot operations work, right? It’s the mix of vendors, uh, personnel skills needed to operate that smoothly and at scale. All of that is sensitive too. Um, and. Unfortunately, there’s this kind of like paradox of collaboration, right? You have to be open enough to collaborate and like give something to the partnership, but you can’t reveal all the goodies. And I think that what we’re seeing is folk understanding that, oh, you can’t reveal all the goodies. You have to be a little careful about it.

Grayson Brulte: So what is the IP that you’re seeing in they’re most concerned about? Is it. How, how the system operates from And Tom strike. Yeah. Is it how operations go? Is it financing? What is the ip or is it supply chain? That’s a, that’s a really big one. What, what, what have you seen there that, that you need to, that you’re watching for, that are showing up in the, in this deal flow?

Hugh Nguyen: Well, I, I, I can tell you what I don’t think it is, uh, before I tell you what I’ve come to realize, it is, um, you know, from a, from a. Technology, like patent perspective, it’s, it’s not that, um, as robax operations scale and as many different kinds of players who have historically never played in this space. Rental car companies right, for example, come into it. There are a lot of basic questions asked in the realm of safety, safety management, fleet operations, branding, um, a lot of questions that. Companies that didn’t kind of grow up in this space, in this small town, right. Um, are are naturally asking, um, you know, industry to industry safety management standards differ. Um, you know, from a Reagan compliance perspective, right? There’s, there’s AV company specific regulation that not every company that is a partner of an AV company now has had to comply with historically. So it’s that legacy that’s that knowledge, right, of the questions being answered now, written down now. In the form of standard operating procedures standards, right? Um, you know, the internal knowledge bases, right? That’s actually what I think is the secret sauce that maybe inadvertently is sort of, um, being developed with a partner and kind of left on the table for anyone to look at and take, but is incredibly, incredibly valuable to protect because it’s proprietary to each and every individual participant in a partnership.

Grayson Brulte: So you, you, you’re right. But if you look at rental car industry, it’s an older, more established industry. If you look at fleet operations, all established, if you look at the finance or debt financing, we can go down lines, more established industries. The cultures are, whoa, boy. Oh boy. The cultures are different. Yeah. How do you, when you’re putting together two, how do you get the cultures to mash? Well,

Hugh Nguyen: you have to take advantage of the fact that for a lot of these players, it’s their very first time working with a. A technology provider or a partner that’s in a totally different space and mindset. Um, and when you do that, entering into a partnership with someone who comes from a totally, completely different background, uh, different corporate structure, right? Um, different longevity at that organization or a company. Um, you know, I find that the, the partners that succeed the most are the ones who, despite not sharing common industry background or difference, have a natural curiosity and an understanding to want to know. How to make it work. Right. And a willingness to put those differences, um, and processes on paper in a way that like your partner can understand and to mutually resolve that it’s, it’s not rocket science. Right. The partnership success comes down to, you know, at the brass acks operational level, how much collaboration and willingness there is between teams on the ground, in the field, in the lab to actually. Resolve, um, differences and develop common ground and common procedures. It’s, you know, there’s, there’s a lot of sophisticated work, you know, uh, financial analysis, primary research that goes into picking the right partner, but actually making that partnership work over time. Not that there’s been a huge timescale to some of these more recent partnerships, comes down to the ground level team to being unwilling to actually, um, function as one and. That’s not necessarily easy. Um, you know, I think, uh, a lot of my, my work historically has been done in, um, kind of the non ev legacy OE space where the cultures, especially around supplier management, for instance, right, are very, very different, right, than some of our, our, um, our robax leaders, right? And so managing those differences. And understanding, um, you know what the other wants extremely important. And you can only do that if you have worked kind of on both sides.

Grayson Brulte: Well, that that’s your, if you put your secret sauce or your competitive advantage, but before you get there, you have this lawyer and this team likes to do the paper. This way this lawyer and this team likes to paper this way. Yeah. And then sometimes they’ll argue over a capital T or a lowercase t. Sure. Yeah. How, how do you get that together before we even get to the operational level? And when you get to the operational level, certain companies have different, if you wanna call it, uh, playbooks, they like to deploy. Yeah. This is the way that we do it. Yeah. And then this is the way No, no, no. You’re doing it the wrong one. No, you’re, how, how do you resolve the, the capital versus the lowercase?

Hugh Nguyen: Well, you know, I think that issue of agreeing to terms right. Um. Particularly like, you know, defined terms in a purchase agreement and, and coming to harmony with that, or avoiding the contention that comes with that. The stakes are a little lower when it’s just, um, like a partnership or an open innovation agreement versus, you know, um, something where you’re. Changing the ownership structure of an organization like a JV or, or, or, or m and a. Not, not to say that you don’t find extraordinary legal talent behind these partnership agreements and understanding how to structure them properly and that they’re on both sides of a partnership. Um, but you know, there’s, there’s a little less, there’s. Kind of binding right in, in those, in those partnership agreements. At the same time, it’s not to say that as those partnerships go, those questions don’t come up right. Oftentimes, um, you know, when I think about, um, you know, emerging but soon to be successful partnerships that I’ve seen, you know, in the space, it’s when, um, the Regan compliance and legal teams of both companies disagree on a particular issue, but product teams weigh in on the actual. Customer impact and importance of that issue and have the authority to actually say whether it’s truly a gating issue or not. So at the end of the day, you know, it’s cliche, but the, the product and the experience should win. And when you defer to that product expertise to help make the financial, the commercial decision, um, it tends to succeed.

Grayson Brulte: But what happens, and I’ll just use the, the simple term that the audience will understand, joint checking account. Mm-hmm. So you, you put in assets to the joint checking account. I put in assets to the joint checking account and the event of a divorce or a separation. Oh. Oh. That, that gets complicated. And that’s why you have judges that, that figure that out. Yep. But in the case of these partnerships, it’s assets and sometimes it’s cash. Yep. But, but it’s, but it’s assets. How do you untangle the assets Sure. Of a marriage in a complicated relationship?

Hugh Nguyen: Well, I think it’s, um, that’s, it’s a great analogy, right? I think, uh, the, the way I like to look at. A partnership versus, um, like a joint venture. Whereas in the latter, you’ve already said and defined what you’re contributing and there are assets that have changed, have been contributed, or changed ownership that have to then get segregated If it dissolves, uh, you know, in a partnership group, I actually don’t know how often I’m seeing asset ownership, like being spelled out. So to get to the bottom of the question, right, it’s like, hey, you know, for, so for a very expensive, highly tooled, um. Vehicle, Roboto taxi, who’s gonna eat that depreciation charge. I think that a lot of, a lot of the economics of, of these partnerships, especially if they’re gonna last long term. Where you encounter the concept of the end of a life of a vehicle, right? Um, a lot of those mechanics, right, those questions, I actually, you know, don’t see corporate development and partnership teams resolving like upfront. I know we’re early, I know time will tell, and there hasn’t been a partnership announced just yet in the robot robotaxis space where we’ve seen a vehicle go all the way to the end of its of its, uh, of a 10 year life, right? Um, but. You know, it’s, it’s not the most satisfying answer, but I think those are, those are questions that have to be resolved as they come up, as the health of the partnership is known to deteriorate. Um, you know, over time it’s not spelled out upfront.

Grayson Brulte: That’s the depreciation has to be figured out. But then rental cars, if there’s a foreign substance, I’ll use the nice term, a foreign substance. In the vehicle, it, it goes to the, the crushing machine and it gets crushed and it gets put out. Have we solved that? Resolve that, figure that out with Robax and in a partnership, who takes the hit for that?

Hugh Nguyen: Yeah. Um, so, uh, uh, a lot of it. Comes down to, you know, who, who, who is who, who owns and is registered on the vehicle itself. Um, who is responsible for the maintenance, the cleaning, um, the fleet operator, right. Who is responsible, and you’ll remember in many cases, right? Um, it’s not just the announced partner that tends to be responsible for elements of fleet operations. Mm-hmm. Right? Um, you know, many, many of the companies, um, being announced in these partnerships. Either partially own or fully own their own fleet operations and are typically deferring. Right. Um, you know, uh, the, the, the decision, not the liability, but the decision of what to do with that, um, with that harmful substance in car. Um, so I think it’s, it’s, you know, it’s, I I don’t, there have been kinda reported cases, um, of. You know, partners in this ecosystem kind of going like this Yeah. And saying, you know, we need to figure it out. But in a non-public, uh, you know, way, the, the key thing is, you know, whenever there’s regulatory authority involved, whenever you know the police are involved, that there’s a very clear person on first base. Right. And, um, you know, that that’s spelled out very clearly in agreement between, um, you know, the, uh, the, the partners. And oftentimes it has to be, um, you know, when we think about. You know, a fleet deployment, um, incident response, event response and like, uh, the chain of communication, the timeliness of it to comply with, um, you know, reporting requirements specifically as it relates to AV actually requires both like the hardware operator and the fleet operator and whoever is doing the, like, remote response to actually be in sync communicating very, very quickly. So oftentimes everyone’s involved when that incident happens.

Grayson Brulte: If you look at it from a a, a professional athlete, it’s practice practice, practice, practice. Yeah. And there’s baseballs win the, hopefully win the World Series. Yeah. It’s practice, practice, practice to win the World Series. For those, those incidents, is there a, is there a, a playbook that’s followed or if you wanna use the term a, a head coach? Mm-hmm. They kind of walk us team through it and they, and they practice these scenarios. So if an incident does happen, all the parties are, are ready to come together to implement.

Hugh Nguyen: I would say by and large, yes. Um, I, I, when I think about much of the investment from a non-technology perspective, uh, that roboto taxi players have made in operationalizing a fleet and making it safe for the sake of safety. Scenario as simulation training, the actual live involvement of local authorities in the markets that you’re deploying to and the documentation of those processes, that’s table stakes, right? Um, you know, I think, and I actually think it’ll remain that way because we’re not at a level of, you know, um. Like, uh, we’re not at a level of, uh, like governmental recognition of repetition of like robot taxi deployment that it’s like natural just yet. So you’re gonna continue to see, um, like very methodical, ideally, right? Very methodical, safety centric, collaborative rollouts involving authorities involving robax companies. And you’re gonna see it captured and documented in rigorous. Uh, you know, rigorous sets of documentation, operating procedures that ultimately in some cases have to be co-authored, you know, between partners, right? So, yeah, that’s what I think,

Grayson Brulte: because I use the, the Hollywood term above the line. So that’s your Tom Cruise, that’s your, that’s your movie stars. Mm-hmm. And below the line is the individuals that we’re recording this in this beautiful studio here that, that do that. Why is the focus now is always on? Vehicle count. I harp on it all the time. Yeah, sure. So I’m going through it. It’s vehicle count, it’s it’s partnerships, it’s, it’s the California Public Utilities Commission. Yeah. The, the miles driven. Why is there not this below the line larger focus when I could sit here and make a very intellectual argument to you? Yeah. That 75% to 80% of the, of the deployment goes below the line.

Hugh Nguyen: Yeah. I would actually argue to you in the same vein that what goes, as you put it below the line, is as big of a, um, scaling factor or limiter, right. As. Vehicle supply, right. Or, you know, supply chain constraints, um, or, you know, labor constraints, right? Um, or reg for that matter. Right? Um, I think that the value that some of these early entrants, early expanding partners into markets have is that they’re figuring these reps out for the first time and are the first to understand and discover what the challenges of deploying in new environments are when you’re working with other third parties. Right. And in many cases with these partnerships, it’s not one-on-one. It’s, you know, the, the provider, it’s the fleet operator, it’s the network operator, the provider of the app, right? And then everyone below that as well as the subcontractors and the vendors working for the fleet provider, right? I mean the, the, the network goes pretty deep. Um, you know, it’s that, that, that layer below the iceberg, below the tip of the iceberg right, is really where I think the success. And scaling ability of partnerships in particular, you know, um, is for certain. Um, and that requires equal focus for each new partner that comes into the fold. So for organizations that are pursuing, like multivariate multi-partner strategies, right? Um, you have to be willing to kind of put the rigor in to repeat that for each partner that you go with. And there’s a real cost to that. Right, because a lot of it’s gonna be proprietary partner to partner. And if you kind of go along with what I said just earlier, you wanna be more careful about what you share. It doesn’t necessarily incentivize you to recycle, rinse, and repeat. You know, your fleet ops manual across every partner that you have. So you have to be careful about that. It’s a fine, fine balance.

Grayson Brulte: So m and a, when do we see some company backed by private equity roll up below the line and when they roll it up, they also have the credit facility and the cash. To put vehicles on the balance sheet. Mm-hmm. They have an in-house economist to understand e energy prices and they, they have the access to tap into energy. Mm-hmm. And you wanna call it the whole below the line package? Yeah. When does somebody roll this up and, and basically consolidate the market? ’cause yes, there’s a variety of companies today that are, that are out there doing this. Yep. But. I just think if somebody did it, then we could scale much, much faster.

Hugh Nguyen: Yeah. And I think the market is onto something. I think it’s important to note in that, that what is below the line tends to be operations of sorts of businesses that are not typically as capital intensive as building a self-driving stack. So these are businesses that throw off more cash, right. Have longer operating history. Um, scale. Um, inevitably they are real estate or real estate type businesses, right? Um, that are more mature, very ripe for consolidation. And I think what you’re seeing, and I, you know, I think there are, there are mobility network operators, ride share operators that actually have stakes or fully own, you know, fleet management companies. Um, you are seeing some of these early entrant potential winners actually take ownership stakes. Those operations themselves rather than maybe outsourcing it or, you know, relying on a subcontractor arrangement, they wanna control that experience. There have been a couple of, uh, of, of players companies that, you know, didn’t exist three or four years ago that are now serving some of the biggest names in multiple markets, not just here in the US but globally. Right. Um, so I think there’s tremendous opportunity there. Um, and certainly anyone that is not necessarily looking to roll up, but looking to control the cost and quality of. Complete operations going forward needs to consider ownership stake.

Grayson Brulte: But why invest in multiple? That’s, that’s what I don’t understand. So, okay. So you invest in one, you invest in two, and why not just invest in one and pick your horse and pack? Sure.

Hugh Nguyen: Well, I think, uh, you know, real estate footprint owned by each of the particular players is. Uh, you know, is, is one, you know, factor, right? Um, and I think just thinking about in the case of rental car operators, right, that tend to have foot footprints, um, at high volume, high traffic locations, high turnover locations like airports, um, you know, near civic centers, that sort of thing, that’s a very, you know, different footprint that may not necessarily be suited to say large depot. Operation, right? Mm-hmm. Um, than say, uh, a larger facility, you know, offsite or away from those sorts of locations. So I think the diversity and kind of where these players play physically is one big factor on, you know, why one just doesn’t gobble up, you know, all of the others as well. Um, you know, cost of capital is still being high. You know, aside, I also think that like the, just, just purely from a. From a, from a volume perspective, I don’t think anyone doubts that TAM is there, but there isn’t a ton of historical data about the profitability of these sorts of operations just yet, which is what you sort of need to trigger some of the traditional. Capital and funding, um, that you would need to really force a serious roll up. So I think those are a couple reasons why you haven’t seen that. Um, consolidation below the iceberg or below the lines, you put it right, happen, happened as quickly. There’s a lot of big things

Grayson Brulte: below the iceberg. The poor Titanic one because below the, below the iceberg. So you’re saying if I’m getting this right, we have to wait for maturity to enter the market because. It the last two to three years, you even say four years, you’ve seen roll ups in some interesting businesses. Yep. The landscape business, the HVAC business. Yep. But mature businesses, they have a common narrative. They spit off massive free cash flow. Yep. Yep. So do you think, so 2027, and you and I both agree, is gonna be a very pivotal year. So maybe, are we looking to Q2 Q3, potentially Q4, 2028 where somebody rolls it up and then tries to float it as an AV reit. Depending on the tax structure,

Hugh Nguyen: that would be really interesting. And by the way, before I answer that, I’m gonna change the analogy from Iceberg and Titanic to something just a little more positive. That doesn’t imply tanking. So we’re gonna, we’re gonna, we’re gonna go back to below the line and above the line. ’cause I like that a little, a little better. It’s less of a hot take. Um, you know, I think, uh, so the question being, when do we see the current ecosystem and mode of partnership, um, and collaboration turning into, you know, transaction, um. I think that I’m not gonna throw an over under of dates out there. You and I were talking about the late 27, early 28 timeframe, but my perspective is it’s actually gonna be a little more insidious than just like headline m and a X buys y consolidation as we’ve seen in history in this particular space. The, um, transitions and the consolidations don’t take the form of like big headline deals. It’s strategic acquihires. Right, it’s IP being bought and folded in, you know, by by someone holding it. So I think you’ll see quite a lot of that. I mean, that kind of makes sense because between now and then, even with all the capital flowing into the space. Um, what was true five years ago that, you know, everyone knew each other in this space. Small town feel is still largely true if I look at the audience of the podcast, right? I mean, it’s still largely true. Will that still be the case in 2020? I actually think so. And because of that, you may not see so much adversarial m and a happening. It’s gonna be, you know, poaching this individual, poaching this leader. Right? Um, team is moving from left to right. Um, that’s what I think you’ll end up seeing.

Grayson Brulte: Then do you see well-funded startups with billions of dollars in cash, multi-billion dollars of access to credit to buy some of their smaller arrivals from a talent acquisition presenter, but also to kind of solidify the position in the market.

Hugh Nguyen: I, I think opportunistically, they have to be doing that. I think that, um, anyone with, um, you know, anyone with enough on the balance sheet to be able to fund strategic acquisitions of capabilities or teams. Talent, right? Mm-hmm. Needs to be pursuing that strategy like really aggressively and not just again, from. Like the, the typical engineering and product talent needed to develop and maintain a self-driving stack, but from an operations and a safety perspective, that talent as it relates specifically to AV deployment at scale, very, very rare. Right?

Grayson Brulte: Then we debated this today on, on the panel. How many stacks are there at the end of the day? Do you have, obviously you’re gonna have a specialized stack for, for mining? For off-road. Yeah. For, for ag, for, but then you also have licensing and this.

Hugh Nguyen: Yeah.

Grayson Brulte: At the end of the day, how many stacks can the market really truly support?

Hugh Nguyen: A bit of a non-answer, but when we think about, you know, I, I think there was, there was an earlier question about what model, um, of monetization do we think will win? Is it licensing of a stack, right. Or is it kind of fully owned model? Um. You know, I, I don’t know if we are seeing enough, um, AVS deployed at scale to suggest that we know the top of the tam. Right? Um, and we are seeing companies that have, um, strong investment in one way, say, robotaxis, kind of pivot and kind of go all in right on, on on. Personally owned. Um, it’s really, really exciting to see that companies are willing to do that right now because, you know, there have been some failures in the past. Right. Um, but I think you’ll see two patterns. I think there is room for both personally owned and I, for, for personally owned Robo Taxii and, um, from a personally owned perspective, the licensing model as opposed to fluid up, I think, you know, there’s room for, for, for all of that as long as there are, um, different kinds of passengers, different kinds of use cases. Um, you know, as long as there is a service industry and people using cars for, you know, kind of personal routine, right? Different frequency, different need, there will always be, uh, a market and a case for, for varying levels of autonomy, capability. Um, and not every OE wants to, or is going to even despite announcements, right? Um, make that commitment or investment to replicate, um, getting to an L two plus capability when someone else can clearly do it. Um, so I do think you’ll see room for that model to succeed. You know, as, uh, as well, I, I don’t see a major consolidation of capabilities or prevailing winner, you know, in that space.

Grayson Brulte: Now we’re gonna put another ingredient here. What happens when the passport model enters the picture? And what I mean by the passport model is you buy a vehicle, you buy an L four person autonomous vehicle. Mm-hmm. And they’re, they’re gonna have restrain times not gonna work everywhere. It’s not level five. Yeah. Then, so I’m in Florida, we come out to California and then I can get into a robo taxi. What happens when that model comes in? Is that gonna be partnership model? Yeah. Where if you buy this vehicle, you get access to Unlimit Unlimited rise in this Robo taxi, or do you see it’s the Tesla model? Yep. Do you see Waymo going that following the Tesla model do, do you see other companies following that model where now you’re locked into the ecosystem? It goes back to Apple?

Hugh Nguyen: Yeah.

Grayson Brulte: If you’re locked into Apple photos and ims, you can’t get out of the ecosystem.

Hugh Nguyen: Yeah, that’s exactly right. I mean, like, if this evolves into there, there’s, there’s been this notion of mobility as a service for, you know, a couple decades, right? Mm-hmm. Um, and, and we saw it take shape in the form of, you know, ride share. And we’re now seeing another flavor of it take shape in the form of, like autonomous rideshare. And as long as there is, um. As long as there are like dominant major players that are early in the space as it grows, there will be an incentive for them to create ecosystem place, right? There will be an incentive for them to partner with other partners and providers. Um, you know, to be very sticky in the audiences that they work with. I, I don’t know if any individual robax provider right now is as deployed globally in as many markets to have all the information and the massive volume and data needed to know how sticky you know they are. Right? I mean, I guess in San Francisco you can clearly see the performance of. Like, you know, uh, an AV offered by Waymo versus, you know, not offered by, by Uber and by Lyft, right? But I, I, I don’t know if enough information in enough different kinds of markets globally exists yet as all these deployments are happening, for the most part, at least, you know, not counting my mobility and, and they’re, they’re very apt strategy of looking at smaller communities, right? Um, like all these major markets being announced are, you know, large. Metro. Um, I, I just don’t think you see enough data in diverse enough markets to really, um, you know, make that call.

Grayson Brulte: Yeah, that’s interesting. May may make out the grab deal. Yeah. Which is really interesting. Yeah. If you look at, ’cause I believe that the global OEMs are gonna have to license. Yeah. There, there are less than one handful mm-hmm. Of global OEMs. I think they could actually pull it off in-house and the other 99.9% are gonna have to license. So you do this, so let’s just say this automotive company licensed this stack. Mm-hmm. But then that stack’s also in this robax. Could that be a way perhaps that the, that the partnership works because there is a, a common licenser or technology on the back stack.

Hugh Nguyen: Interesting. So kind of a pass through based on, yes. Yeah. Based on who else is a customer? Who else is a licensee of that licensor Right. To do it. I, I think companies that. For instance, if you are, if you’ve kind of made your bed in the, I’m not gonna play in the Roboto TAXII game. I’m gonna kind of stay purely focused on licensing out a eight as capability, you know, for my car. But by virtue of who you choose to license with, you have access to that knowledge and that intel concerning how Robotaxis operates. I mean, you’d be a fool not to use that information and that knowledge, right? So I think you will. You will see that. You know, pre, pre pandemic as you know, right. Um, there were a number of transactions including transactions that, that, that I was involved in, um, where you saw, um, OES making bids to acquire and consolidate software talent for the purpose of, you know, building inhouse capabilities, right. To, um, make all of these, um, autonomy dreams eventually come true in, in whatever form. Uh, I, I don’t see. That talent concentration, that poaching happening as obviously Right. Um, I do see it happening on a smaller scale and I do think that, um, you know, that trading of talent, um, will determine which license product gets the most licensees and ultimately ends up winning and subsuming others.

Grayson Brulte: So if you wanna use the term. Precision strike. Mm-hmm. For example, so you get this, this rockstar of an engineer or, or this individual executive who’s a really great engineer. That’s interesting. Yeah. And so that’s where you think this is gonna go with precision strikes.

Hugh Nguyen: Um, precision strike specifically around who can help me generate the highest quality and most relevant synthetic dataset, you know, for training, who can help me get to unique edge case identification much, much faster. Right. Um, who can help me with the management of all of that data that I’m creating, right? Um, those are the sorts of skills and talent that I think will be surgically or I guess precisely, you know, gotten. And it’s not gonna be through these like big corporate development driven wind up m and a processes, right? It’s gonna be through like, again, small town leaders who have credibility having. Conversations with Fcon industry, sort of making that happen company to company. Um, which when you think about partnerships, the original theme, right, we’re kind of coming back to right, that’s kind of one of the best parts of, of these partnerships in the sense that if they’re non-committal, you’re getting access to understanding your partner’s talent. Where they’re strong, where you’re not in a way that you wouldn’t have if you hadn’t entered into that to begin with. So in the paradox of like collaborating, it’s almost worth that risk just to understand what capabilities you don’t have, particularly as it relates to talent,

Grayson Brulte: but how do you mitigate that risk? Yeah.

Hugh Nguyen: Yeah. It’s, it’s, it’s tough. It’s, it’s long. You can just go out there, right, and pick someone off the street and say, Hey, you know, build me a synthetic data set, right? Um, it’s, it’s, it’s tough. Well, first you have to. You have to train it. We, we need, we need more individuals, more practitioners in that pipeline, developing those capabilities and, and, and training it. Um, and I think there are some amazing organizations, um, in that space that have all, you know, been guests of yours that. Beyond contributing to the ecosystem by licensing or being partners with players in the ecosystem, are contributing to the ecosystem by developing very unique AV geared talent, right? That you can’t develop without experience. And so I think we need to see many, many more of that, right? If I think about, you know, you know, the self-driving project and how it sort of spawned every, you know, many, many executives right in the industry who’ve gone on to lead their own efforts, right? I think you see miniature versions of that happening with some of the. Players that are kind of sneakily ubiquitous today, right? Particularly in the training game. So I see that happening as well. It’s not necessarily a litigant, but I guess there really is no MIT when the supply of that talent is, you know, still so short. The question is when we start to scale, right, um, you know, that becomes a real, real constraint. Then again, scaling requires a different commercial and operational skillset that I think ideally, hopefully, is gonna be more available.

Grayson Brulte: The thing that’s, I would say it is, it’s dwindling this time versus the last big cycle was the cult of the founder. Mm-hmm. With certain individuals or could do no wrong, could raise massive amounts of money. Yep. When do we kind of drift out, out of that to just pure operational excellence is what everybody’s talking about. Not necessarily, yeah. Who the founder is.

Hugh Nguyen: Uh, you know, I. As a data-driven person, um, working for a firm that really prides, um, kind of looking closely at, you know, what the numbers say and what they infer, um, it can be frustrating in this space because. You know, other than, you know, um, cost per ride mm-hmm. You know, miles driven, miles logged incidents per a hundred million miles. There aren’t a lot of hard metrics that you can measure performance on. Like in this space, in in particular, even like the unit economics, right. Um, you’ll, you’ll be on Reddit, you’ll see a lot of different. User made comparisons, right, of what they think the unit economics are between like a robo taxii, right, deployed by X company versus, um, like a regular, you know, individually owned ride share. There just isn’t a lot of very credible data that’s third party generated that doesn’t come from one of these companies, right? So absent of that data, which will solve itself over time as deployments become. Sure. Um, and as time passes and these partnerships have a chance to succeed, right, that will pass. But in this day and age, right here, right now, measuring the effectiveness of one of these deals, not a lot of hard data behind it. Certainly, you know, not profit data.

Grayson Brulte: There’s not a lot of data, but we look to wrap up. What is your outlook for m and a over the next 12 to 24 months in autonomy?

Hugh Nguyen: Um. I think that the current mode of dating, speed dating may be partnerships, open innovation, autonom speed dating, please. Exactly. If we say, say as it were, as it were, right. Um, yeah. You know, that will, that will certainly continue. Um, you know, if we look at the world as a chess board and we look at each major city as a market, um, for, you know, autonomy with its own. Regulatory, logistical challenges, supply challenges. Much of the world has yet to be announced yet, right? There are still many major metropolitan areas out there, um, that are ripe. Um, uh, customer bases for, for robax use cases that, you know, haven’t popped up just yet. Um, so we’re very early right in that game and as we’re early in that game, there still is room for alliances to be formed before people, you know, get kicked off the island and, and, and don’t succeed. So I still think you’re gonna see a great 24, call it 18 to 24 months of. The rash of partnerships that you and I have noticed, right, being announced. And again, when some of these partnerships do not pan out the way one or other partner want you will not see it broadcasted you, will not see it, um, you know, proclaimed anywhere, right? You won’t see, um, you know, earnings call commentary about it, right? So you have to look very closely, uh, to know who actually is. Losing. Right. And then make judgements on who the winners and the losers are there and the winners will consolidate. Right. Those that are traditionally heavily capitalized, well-funded have at least, you know, investor sentiment behind them. Right. And very, very importantly, and I’m gonna double down on this since it’s your show, a handle on how to shape the policy and the markets that they’re in. While they can, those will be the winners. And those will be the acquirers.

Grayson Brulte: Yeah. I mean, the world is a big place. There’s a lot of to stuff to analyze. That’s what we do each and every week here on the road to autonomy. Hugh, it was great to have you on. He’s coming back. The future is bright. The future Thomas. The future is a growing and thriving autonomy economy. Thank you so much, sir. Thanks Grayson.