Arbe Robotics Ltd. (ARBE) Earnings Call Transcript
May 28, 2026
Earnings Call Speaker Segments
Thanks, and hello everyone. Welcome to Arbe's conference call, where we'll be going over our results for Q1 of 2026. Before we get started, I want to remind those listening that some of what's said on this call could include forward-looking statements, and the safe harbor statement that's laid out in today's earnings release also applies here. If you haven't gotten a copy of that release yet, you can find it in the Investor Relations section on the company's website. Joining us today is Ram Machness, our Chief Executive Officer at Arbe, and he'll kick things off with an update on the business and strategy; Kobi Marenko, our President and Co-Founder, who will share his perspective on where Arbe is headed over the long term; and Karine Pinto-Flomenboim, our Chief Financial Officer, who will go through the financials. Once management finishes its formal remarks, we'll open up the call for the Q&A portion; so with that, let me turn things over to Ram. Ram, the floor is yours.
Thank you, Kenny. Good morning to all of you, and I appreciate you being here as we go over Arbe's results for the first quarter of 2026. I took on the CEO position in April, and I believe there is considerable opportunity in front of Arbe. We achieved solid progress during Q1 and moved faster on the plan we laid out when the year began. Today, my focus will be on the key areas where progress is happening. Arbe is evolving from a company centered on chipsets for automotive into a provider of full radar solutions spanning automotive and, increasingly, adjacent markets. Several commercial milestones were reached this quarter that show this shift. We started delivering chips into China via Hirain, our Tier 1 partner. Orders came in from robotaxi customers; we also got orders for data collection programs involving global automakers and top mobility players. Now I'll go into more detail on each area, beginning with automotive. Let's begin with China, the automotive market moving fastest anywhere in the world. This quarter, we delivered the first batch of chipsets to Hirain; those chipsets enable production of 48x48 channel radar, a project announced in December where Hirain is building a Level 4 autonomous vehicle solution for a Chinese automaker. At the same time, Hirain is working on new radar, also built on our chipsets, with 24x12 channels, intended as a lower cost system usable across a broader range of vehicles. That configuration pairs with its current high-end 48x48 radar system powered by our chipset. Through our technology, Hirain is emerging as a major radar platform player in China. It can give OEMs a route from high-end radar systems available today in China all the way to full 2K ultra-high-resolution performance. China sold 34.4 million vehicles in 2025 through a local Tier 1, and Arbe obtained direct access to one of the biggest and fastest-moving markets globally for Level 2+ and Level 3 deployments. Moving to robotaxis. Orders came in for our Phoenix radar system from robotaxi companies worldwide. These systems enable Level 4 autonomy and full 360-degree sensing. This makes it clear that our high-resolution radar provides the performance and coverage Level 4 demands. More broadly, we observe strong and expanding interest in our automotive chipsets. We are involved in data collection programs with global automakers and leading mobility players. We have moved into advanced collection processes with certain Chinese and European automakers; this evaluation increasingly centers on the specific use cases where cameras and LiDARs come up short and where imaging radar becomes the essential sensor. This reinforces industry interest in our radar technology as a core sensing platform for autonomous driving programs; that interest arrives alongside a significant shift in the automotive market. Lately, a number of leading OEMs have taken another look at their Level 3 programs. We don't view this as a rejection of eyes-off autonomy. We view it as a reset, returning to the proper fundamentals of autonomy. The first-generation Level 3 systems had obvious limitations. They were geofenced. A number of them had constraints on speed, others could only function in favorable weather conditions, and plenty of them were never commercially launched at all. In addition to how mature the algorithm was, we think another key factor behind this involves sensor performance—more specifically, imaging radar failing to deliver the performance required. From what we've heard in conversations with numerous OEMs, automakers today are actively evaluating next-generation eyes-off platforms and seeking sensing capable of supporting Level 3 use cases; that gap is precisely what our high-resolution radar was built to address; we also set out to learn what drivers actually expect from eyes-off autonomy. A poll of 1,000 individuals spanning the U.S., Europe and Asia delivered an unambiguous signal: drivers would be willing to pay, to change their vehicle and even their brand in exchange for fully operational Level 3, Level 4 autonomous driving. Consumers stand ready for eyes-off autonomy, though only under conditions of safety, smoothness and reliability. Arbe's ultra-high resolution radar delivers precisely that capability. We welcome you to visit our website for further detail on this survey. Before I move past automotive, allow me to say a few words about physical AI, since it's growing increasingly central to where our technology is deployed; we're witnessing how the AI and machine learning revolution is bringing forward a new generation of algorithms; those algorithms, and in particular vision language action models, are reshaping the way intelligent systems perceive, comprehend and engage with the physical world. Within automotive, it produces nothing short of breakthrough progress for Level 3 and Level 4 capabilities. And outside automotive, it unlocks a broader opportunity spanning robotics, logistics and other autonomous systems; such physical AI systems can only be as effective as the real-world data they take in, particularly in safety critical settings like vehicles; that's precisely where we fit best. Our radar delivers dense sensing, long range, low latency and consistent performance regardless of weather. That's exactly the kind of machine understandable input that AI-driven autonomy requires. In the quarter, NVIDIA announced it broadened the global DRIVE Hyperion ecosystem to speed the road to full autonomy and named Arbe as part of its platform. Our continuing collaboration with NVIDIA on radar-based free space mapping and AI-driven automotive capabilities stands as one illustration of how we're positioning Arbe at the heart of this major shift. Now let's turn past automotive, where our technology is unlocking meaningful new opportunities. The first significant development is that Arbe has started selling complete end-to-end radar systems alongside chipsets. This lets us serve markets where customers want a full radar solution and where sales cycles run much shorter. We've already begun shipping these systems to players spanning defense, homeland security, transportation applications, perimeter security, physical AI and several other applications. To back this up, we've also established dedicated production lines to scale system manufacturing. Considered together, these milestones demonstrate that our leadership in automotive radar can carry over into perception-critical markets; we're extending our reach, widening our opportunities and substantially growing our total addressable market. Having said that, I'll now hand things over to Kobi Marenko, our President and Co-Founder.
Thanks, Ram, and hello everybody. Now that I've taken on the President position, my attention is fixed on Arbe's strategy for the long haul, on our key partnerships, and on steering the investments that speed up Arbe's coming stage of expansion and commercialization. In Q1, we made our balance sheet stronger. We completed an underwritten registered direct offering, bringing in $18.5 million in gross proceeds. That provided the financial room we needed to put money into Arbe's present period of rapid growth. What this quarter shows us is what makes me quite confident. Arbe's technology platform reaches wider and can target several end markets in addition to automotive. The very same ultra-high-resolution radar we created for automotive can be — and now is starting to be — taken up on a much wider basis. As Ram already noted, we're seeing more interest and more potential, and in certain cases initial system sales, in homeland security and smart infrastructure. And last, the emergence of physical AI, along with what OEMs need for improved autonomy solutions, matches our strengths perfectly: high-definition, long-range sensing of the environment in all conditions. Ram recently took on the CEO position, bringing a strong focus on execution, deep product knowledge, and the commercial understanding that's required. That will carry Arbe from where we stand now to becoming a scaled radar systems company spanning many markets. I've collaborated closely with Ram across many years, and I have complete faith in what he can do, as well as in the whole team, to bring Arbe into its next phase. My excitement about Arbe's potential over the quarters and years ahead has never been greater. With that, I'll hand the call to Karine so she can go over the financials.
Thanks, Kobi, and good morning, everybody. I'd like to walk through our financial results for Q1 2026 in greater depth. First quarter 2026 revenue came to $0.5 million, versus $0.4 million during Q1 2025. Our backlog stood at $1 million as of March 31, 2026. Gross profit in Q1 2026 was negative $0.1 million, compared with negative $0.3 million a year earlier. Now on to operating expenses. For Q1 2026, total operating expenses reached $11.2 million, a decline from $13.1 million in Q1 of 2025. That drop in operating expenses stemmed mainly from reduced share-based compensation expenses, reflecting earlier grants that are now fully vested, together with last year's award being structured with half in cash and half in equity. The decrease also stemmed from the Q1 '25 tape-out expenses, which have come down as we move closer to productization. Partially offsetting this reduction in operating expenses was the weakening of the U.S. dollar, mainly against the Israeli shekel, and to a lesser extent, labor-based provisions and a merit increase. Our operating loss for the first quarter of 2026 was $11.3 million, versus an operating loss of $13.4 million in the first quarter of 2025. As we announced previously, during the quarter we put cost reduction measures in place that should lower our ongoing operating expenses by roughly 15%, with the full impact anticipated to begin during Q2. Adjusted EBITDA, a non-GAAP measurement that excludes expenses for noncash share-based compensation and for nonrecurring items, showed a loss of $9.9 million in Q1 of 2026, compared to a loss of $9.7 million in the first quarter of 2025. In management's evaluation of our use of cash, and in planning and evaluating our cash requirements for the coming period, we believe this non-GAAP measurement matters. Net loss for the first quarter of 2026 came to $9.4 million, compared to a net loss of $13.8 million in the first quarter of 2025. Net loss in Q1 2026 included $1.9 million in financial income, versus $0.5 million of financial expenses in Q1 of 2025. As of March 31, 2026, Arbe held $53.6 million in cash and cash equivalents and short-term bank deposits. Now to our outlook. The 2026 guidance we provided in February '26 is being reaffirmed. We still expect full year revenue in the range of $4 million to $6 million, and an adjusted EBITDA loss in the range of $28 million to $31 million loss. At this point we'll be glad to take your questions. Operator?
I was hoping you could expand a bit further on the momentum you're observing when it comes to defense-related use cases, drones, and similar areas.
George, thank you for asking that. Essentially, when it comes to defense, our radar is capable of serving a handful of applications. The first one isn't all that different from what we do in automotive; that means autonomous vehicles, be they supply trucks or armored vehicles too. Every one of them needs to travel safely, particularly off-road where conditions are thick with dust, rain, fog, and every major challenge that we — that radar happens to be the optimal sensor for. And on that front, we already have a customer, the American Army, which purchased hundreds of units from us; and we foresee another — we think we'll land an additional order there too. Application number two is defense — protecting perimeters, whether against human beings attempting to enter some area of a perimeter we wish to protect, or even drones capable of attacking or dropping supplies or drugs into those zones. And we certainly view this as an application our radar is able to serve. Beyond that, though, there is — I'd say those represent the two primary initiatives we're engaged in, yet there are several other smaller applications where radar, and our radar in particular, can be of help. And whereas a year ago, we weren't even thinking about those markets, now we're concentrating on them; and I'm working to — we're working to pursue business development alongside every leading player, the major names within the defense sector, along with the newcomers. Additionally, this market is undergoing a genuine transformation. The players present 10 years ago, such as the Lockheed Martins of the world, are evolving too, while new emerging players are capturing their market share; and we're striving to collaborate with the established players as well as the new ones, delivering our solution.
And as a quick follow-up, you brought up the NVIDIA reference design, of course. What I'd like to know is how those discussions have gone, and how much extra momentum has come from it? Or in what ways has it shifted the talks you're having with OEMs and other partners about putting your imaging radar to work together with NVIDIA?
Yes. So my sense is that this question is really about the — which companies are actually in a position to potentially deliver an AV stack, meaning the autonomous vehicle software stack, to the market, since at the end of the day, software is what autonomous driving comes down to, along with algorithms and machine learning. And without question, NVIDIA sits at the top among market players on that today. So that matters a great deal from every angle for all the OEMs, whether they've chosen NVIDIA or haven't made that choice yet. My sense is that all of them are watching what's happening across the industry and asking who can actually get the AV stack across the finish line and genuinely deliver a reliable, sustainable solution to the consumer; so in the end, what you want is to get into your car in the morning and travel safely to work while the car handles the driving. That way you can do something else and reclaim your time by making use of your car's autonomous vehicle capability. And a race is underway at this moment. A lot of companies are pushing toward that. Without a doubt, NVIDIA is likely the strongest competitor out there. So what happens with the NVIDIA platform and the NVIDIA AV stack is extremely, extremely significant to everybody, no question. Other companies exist too. Beyond that, certain OEMs are attempting to develop their own in-house stack. So those players are in the mix as well; but everyone is watching what everybody else is up to, and there is unquestionably a distinct race happening on that front right now. What strikes me as genuinely significant is the progress we've witnessed lately at the algorithm level, plus the breakthrough that all of us encounter in daily life with deep learning and AI. That also carries over into autonomous vehicle software, and we'll watch it arrive in the market quite soon, beginning in China and then advancing into the Western world.
I've got a few things I wanted to ask. Could you walk us through — from what it sounds like, there are several robotaxi possibilities on your radar. Do those lean toward a particular region? Or do they span the globe, that is, North America, Europe, and China? Or is it just one area where they're concentrated?
The chances we're seeing at this moment originate in North America, plus we're also seeing them in China. So without question, the biggest sales come from North America and China.
All right, that sounds great. Then, regarding what you said about delivering a whole radar setup rather than just a chipset, could you perhaps explain the difference in revenue, or maybe the difference in pricing, between those two, when we consider unit expansion for both products?
Sure. So from our perspective, multiple elements play a role in the shift and the substantial rise in AUP—the average unit price—when we're discussing in particular the non-automotive side, meaning non-private automotive, there are two elements. First, the reality that what we're offering is an entire system, not merely the chipset, and that clearly pushes the average unit price higher. On top of that, the volumes are far smaller. So if we're engaged on a Level 3 program alongside an OEM, the volumes are extremely large, correct? Yet in that case, bargaining leverage and economies of scale push the AUP lower. When we're discussing those sorts of programs—whether it's a robotaxi, robot trucks, commercial vehicles, or defense applications—the volumes are considerably smaller on the one hand, yet the AUP is considerably larger, because there's no economy of scale at the level you get within the automotive sector. As a result, we observe a considerably greater AUP in the radar business—the full system—relative to automotive chipsets.
All right. Then here's a minor question: over the long run, do you expect the gross margins on the two product lines to end up alike? Or is it your view that the complete radar system could be lower in the long term?
Yeah. Typically, gross margins on mid-scale projects involving systems come in somewhat below those on chipsets. They're somewhat lower. That's what you'd always anticipate, since with chipsets we're charging for the differentiation we deliver, right? With a system, you're also charging for the mechanics, the chassis, and the other components that are part of it. So if you examine just the COGS margin, they could be the profit, though — the dollar profit for each system is much greater. So on a percentage basis, yes, it's lower. On an absolute dollar basis, it's greater.
Naturally, it's also worth bringing up the volumes, since those vary quite a bit from one system to another and from one chip to another.
Yes. No, I get it. That's great. And then just one final thing from my side, perhaps for Karine. Regarding the backlog, should we understand that as roughly a 12-month figure, or is it shorter or longer?
No, it's 12 months, that's right.
All right. Great; and yes, so — and perhaps just so I can get this straight, that $1 million, is it reasonable to assume it would be used up in '26? Or do you mean it's more like 12 months starting from the close of Q1?
It's 12 months from the close of '21, but my assumption is that all of it will fall in '26 regardless.
What did it cost, I'd like to know, to get the radar's entire production line up and running? And with the capacity you've got right now, I suppose, is that enough, in your view, to handle the defense and AI work already in the pipeline? Or do you figure more capacity will have to be brought on as time passes?
Our production line gets built to serve the capacity required at present, which we take to be hundreds of systems each month, thousands of systems each year. The outlay isn't all that dramatic. It lands on the low side of the hundreds of thousands of dollars, no higher than that. And we're able to raise that capacity as time goes on and as demand expands over the quarters ahead.
We've also kicked off talks with a few contract manufacturers about growing this production line at some point later on.
I want to express gratitude, on behalf of everyone in management here at Arbe, to you, our shareholders, for the interest you keep showing and the backing you give our business over the long haul. This quarter opened a fresh chapter for Arbe — one centered on converting our lead in technology into commercial scale spanning numerous markets. We now have our initial system-level order, robotaxi momentum that is building, a growing opportunity in the Chinese market via Hirain, and an unmistakable spot at the heart of the coming wave of eyes-off autonomy and physical AI. Given all of that, we feel Arbe is positioned well to create meaningful value. We're eager to share our progress with you next quarter, and speaking for myself, I'm excited to step into the CEO position and take Arbe to the next level in its development. That brings our call to a close. Have a good day.
Thanks so much. That wraps up this call, and we appreciate everyone joining us for the presentation. At this point you're free to hang up, and we hope the rest of your day is great.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete Arbe Robotics Ltd. transcript - plus 132,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to Arbe Robotics Ltd. earnings transcripts and 132,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.