Your first question comes from the line of Aaron Rakers from Wells Fargo.

Congrats on the strong results here. I'm curious there was a lot of commentary around 6G and the setting forth, kind of, the standard path as we move forward. As we think about Keysight's participation in 6G, I'm curious of how you would characterize the opportunity relative to the 5G cycle that we saw several years ago. Any kind of framing of when we should expect to see some materializing revenue from a 6G cycle? And any thoughts on how you would frame that relative TAM opportunity versus, let's say, 5G several years ago? And I have a quick follow-up.

Yes. Thank you, Aaron. Yes, it's a great quarter. The team has been executing very well, and we're pleased with that. Relative to 6G, any time you start a new generational cycle, you always look for what's different versus the past. And it's often too -- you can't wait too long to call it. And that's why we've had a focus on making this company about solutions and about first to market. And so we have been engaged with the industry over the last couple of years. And so as I noted in my prepared remarks, we start to see the industry coalescing around early 6G standards in the 2029 time frame. There's the Olympics in the United States. That's the other mile marker. And that's not new. I mean, any time you've had wireless standards evolutions, ironically, it involves a sporting event of some kind. And so that's another mile marker to draw historical parallels. But from a technological standpoint, traditionally, you see new spectrum, and in support of higher speeds and feeds. It's sort of like the base case for any technology, and that is also going to be true in 6G. But as we noted, we're also seeing other vectors of innovation, AI-RAN being one of them, new use cases like ISAC, security infrastructure and also bringing forward the tighter integration of nonterrestrial and terrestrial assets into a communication framework. And all of these are areas we've invested in, and we have the solutions portfolio, and we're working with industry-leading customers. So our base case is the opportunity in 6G is greater than the opportunity that we saw in 5G, and we're well positioned to capitalize on it.

Yes. And then as a quick follow-up on the wireline side. I know you had mentioned that this is the first quarter for which you saw wireline surpass the wireless business. As we think about AI and just the continual expansion of the opportunity set around that, is there any way to kind of help us think about how meaningful AI is to your business today, either within the wireline or in aggregate? And how much of a growth driver that appears to continue to be as we look forward?

Yes. I mean we're very pleased with the wireline business and the pickup we're seeing in the AI-related demand in that business as a primary. Of course, as the opportunities increase, we will start to see secondary opportunities, but we'll try to frame it up for you in terms of our wireline business. Wireline was greater than wireless, not just this quarter, but even year-to-date, we have seen tremendous momentum in our wireline business, and we're very pleased even the strongest quarter for our AI and wireline opportunities in Q3, and the pipeline continues to grow strongly for us. So the way I see it is it's a -- we are in the early stages of a long adoption of AI, not just in wireline and as things come together, we'll start to see a convergence with wireless and convergence with automotive and many other end markets that we're well positioned to capitalize over the long term. But in the near term, very pleased with the traction we're seeing for our differentiated products and solutions, which is far exceeding our ability to supply at this point, and customers are planning ahead, and we're doing very well with our AI business.

Your next question comes from the line of Meta Marshall from Morgan Stanley.

Congrats on the quarter. You mentioned kind of a lot of different ways in which more -- there's more markets or more different types of technology to be testing as far as AI. But could you just give a sense of kind of how testing density has changed? So there's a lot of different -- more end markets, but just kind of how the overall content of testing has changed as kind of some of these technologies get a little bit more complex. And then, Neil, very healthy incremental margins again this quarter. Just any guardrails that we should think about just as we kind of progress forward?

Yes, Meta, as you've noted, as the year has progressed, not only are we picking up what we would call traditional opportunity associated with CapEx investments that customers are making, but the opportunity set as we see it into the future continues to grow and expand. And it's a function of this ecosystem over the last few years has largely been a homogeneous integrated vertical stack with a finite set of opportunities that we have done extremely well. But by working early, we're also working with other players that are entering the space and the space is increasingly becoming more heterogeneous in nature, all the way from compute to racks and also protocols. I mean the -- we're seeing a growth in the number of protocols at all layers of the stack. And the architectures increasingly involve GPUs, CPUs, DPUs mix. And it's not just for the sake of making things more complex. It's -- the reality is different customers have different strategies. And based on the type of workload they're using, they're trying to pick the right architecture for them. And our tools are increasingly doing very well with customers across the broader tail, which positions us well into the future. I don't know, Kailash, if you have any other comments to add.

Yes. Fundamentally, the design margins are shrinking, right? So with higher data rates, lower latency, AI needs to be lossless. And even if there's a limited amount of gap there, the models won't perform. So what our customers are seeing is they can no longer guarantee anything by design. They also need to test it in production as well. So this is increasing a lot of design emulation and test intensity. If you look at a compute or a switch tray these days, it's gone from tens to hundreds of high-speed pinouts. And that's more insertion points for us. We have our VNAs and oscilloscopes testing things at signal level. We introduced a new portfolio to test things at a bit level, and our AI workload emulators are testing things and emulating things at a protocol and packet level. You look at scaling and things are going from monolithic chips to chiplet architectures. So the interoperability of chiplets need to get emulated and tested. Customers are asking when a chip is exercising a model, we emulate an environment for that chip to get stress tested. And they want to see if the chip shuts down or gets overheated, they want to activate all of the cores and with higher power and higher speeds. So all of these are creating additional opportunities for us. And we're excited about the complete portfolio we have, electrical, optical, RF, digital and protocol. We're bringing all of these capabilities to enable our customers. And we're seeing R&D as well as our manufacturing business grow significantly.

Yes. And Meta, to your second question, obviously, we've seen really strong core operating leverage this quarter. And I think as we look forward, I continue, at least as it relates to '27 to feel confident in our ability to continue to outperform our 40% leverage target, particularly given the synergy realization that we'll see. I mentioned that we've largely completed our integration of the recently completed acquisitions. You put the question in the context of guardrails. The only thing I would just caution people to pay attention to is we did have the onetime tariff impacts this year that kind of artificially pulled up '26 profitability that won't repeat. So if you adjust for that and think about it on an operational basis, I would expect we'll continue to outperform the 40% metric.

Your next question comes from the line of Mark Delaney from Goldman Sachs.

Congratulations on the strong results. I was hoping to talk around demand sustainability to start. I think orders have been over $2 billion for 2 quarters in a row now. And as you look into the fourth quarter and next year, do you think this level of demand is sustainable or even a level that Keysight can grow from?

Yes, Mark, we think it is. I'll just say our base case is orders slightly up from Q3, in line with seasonality and then following that seasonal trend into Q1 of '27. Steve, I know the pipeline, you may make some comments there.

Thanks, Satish. Well, let me start by saying it's great to be head of sales at Keysight right now. We had an outstanding Q3. We delivered our highest quarter ever for the third consecutive quarter. And with the traditional uplift that we expect in Q4, we're confident in delivering another record quarter and surpassing $2 billion for the third consecutive quarter. Even more promising, despite these record quarter results, our pipeline has continued to grow throughout the year and now stands at an all-time high. I think this is a proof that our go-to-market strategy and priorities are working. Our top priority has been to really spend more time with customers as much as possible, identifying those new opportunities and find those unarticulated needs and capturing new logos. As a case in point, year-to-date, we've added nearly 3,000 new customers, representing more than $100 million of incremental business. And we're also partnering closely with marketing to broaden our reach and get to those customers earlier in their buying process. In addition, one of our focus has been to elevate our engagement with our top customers. So our largest customers are doing well, performing up high double digits for the year. And more importantly, we're really expanding our reach across their entire ecosystem. We're also adding capacity in markets and high-growth areas. Southeast Asia is a good example. We've more than doubled our business and is our fastest-growing region for the quarter and for the year. So all in all, we further accelerated our momentum this quarter, resulting in our highest ever monthly funnel intake just last month and our new record for rolling 12-month funnel.

Very helpful context. My other question was on supply and the ability to meet this level of demand the company has seen. So can you double-click a bit more on Keysight's ability to meet demand at these types of volumes, both in terms of the supply chain and the ability to get enough parts as well as your own ability from a manufacturing standpoint?

Yes. Thank you, Mark. As you heard from Steve, we're seeing broad-based strength in demand. I think one of the points that in addition to everything Steve said is we're seeing demand across the globe, and our portfolio is doing very well. Now from a supply chain perspective, our team has done a great job this year, continuing to scale with discipline, as you see from our gross margins at record levels as well. And we're continuing to meet our customers' demand needs. as we go through the year. It's also true that the supply environment is less flexible today than, let's say, a year ago. And so we're working with our suppliers to deconstrain the supply chain, especially at these levels of demand, and we remain confident in the guide that we have laid out for Q4.

Your next question comes from the line of Tim Long from Barclays.

I'll ask one then come back with my follow-up. I want to go back to commercial comms and the strong AI business on the wireline side. You talked about some of the applications and the use cases that are helping there. Curious if you can just give us a little update on how that business is looking from an R&D standpoint versus manufacturing. If you can give us a little update on any movements that you've seen in the AI-related business? Then I have a follow-up.

Yes. As it relates to R&D and manufacturing, and I'll talk about it from the wireline side of things where we put out the numbers previously. Historically, this has been a business that's been amongst the most heavily levered towards R&D. We've talked about it at about 80-20. We've more recently said with the addition of manufacturing business in the support of the AI data center build-out that's shifted to more like 70-30. And I think if you look at it over a shorter horizon over the last couple of quarters, it's probably more like 2/3, 1/3. But we're still heavily levered towards R&D within wireline even as we're servicing this entire ecosystem and servicing Keysight's customers as they take products out of R&D into manufacturing and ultimately deploy them into the marketplace.

Okay. Great. Great. And then my second question, my follow-up, there was a mention in there about the hyperscalers. I'm curious if you can just touch maybe at a high level on how important of a customer cohort is that? Are they -- is there a margin difference? Is there more product they take because they're involved in a lot of technologies? And would that be just the 4 or 5 big ones? Or do you see that spreading to the next level of neo-clouds and others playing in the industry?

Yes, I would say, Tim, the hyperscalers and our early engagement started about 5 years ago with our acquisition of Ixia, who had more of a relationship with them in the early days. Right now, I would say our relationship with hyperscalers not only is in the U.S., but I also would add some of the model companies that are driving a lot of the demand for the ecosystem. So it is strategic in that sense because of the tremendous capital that they're deploying and that sort of moves downstream into the ecosystem. So understanding their needs are important. Also, many of them, I think, publicly disclosed their own silicon -- in-house silicon and efforts to make their own chips. So they are a very important customer base for us. From a revenue concentration point of view, they're actually the smaller of the entire, you'd say, roughly 10% of our business is from that cohort of companies directly, but they do have a lot of downstream effect and influence across the ecosystem. So it's very important. What's also important is to recognize that even as we're winning today in R&D and in manufacturing applications that -- where the industry is scaling, we're also uncovering new opportunities such as with regard to emulating the workloads because that's where the future is going to be, and we're in a very good position with our emulation platforms to be able to help the industry to uncover the various heterogeneous emulation of workloads associated with AI because as Kailash mentioned earlier, that is going to be critical, especially given the latency requirements of AI and the scaling data across the AI network.

Your next question comes from the line of Adrienne Colby from Citi.

It's Adrienne for Atif Malik. I was hoping you could talk a little bit more about the sequentially slower growth in aerospace, defense and government segment. You described a lot of strong demand dynamics and double-digit order growth, but we did see a bit of a step down in the growth rate there.

Yes, Adrienne, I've said this, you heard me say this on the call, Aerospace and defense is a business I can easily call years out. Very difficult to call in a given quarter just because it's got government budgets, it moves at its own pace. But this year, we're quite pleased with the growth and adoption of our solutions, including our newly acquired PNT offerings from Spirent. So Neil, any specific comments?

Yes, I would just reiterate that we're still up double digits. There is some quarter-to-quarter perturbations in these end markets. I -- my honest feeling is you're probably -- there's probably nothing to see there from that perspective.

And then just as a follow-up, could you comment on if the run rate that you were seeing in business within wireline is consistent with last quarter? Yes, please.

Yes. Let me just make one more comment that we built backlog in the business. So if you're looking at revenue and drawing that conclusion, that's a function of supply and supply chain. So I would not read too much into it, as Neil mentioned. Please go ahead.

Can you repeat your follow-on question?

Yes. I just wanted to confirm or rather to ask if the run rate of the business, the AI business within the wireline segment was consistent with what you were seeing last quarter or if you've seen that expand at all?

It has expanded.

Your next question comes from the line of Andrew Spinola from UBS.

You reported another strong quarter in the EISG segment. I think you highlighted some of the strength in semi and general electronics. And I wonder if you could unpack that a little bit. You said last quarter, you're seeing some of the demand from AI expand into some of these segments. I'm wondering if you're seeing that and if you think that there's a fairly meaningful expansion still ahead in those segments. And I'd also like you to comment on the operating margin, which was quite strong in EISG in the quarter and how you're thinking about the sustainability of that going forward?

Yes, Andrew, pleased with the double-digit growth we're seeing in our EISG business year-to-date. Recovery in auto is another theme along with strength in semi. But we have Jason here, and he'll touch upon those points.

Yes. Thank you, Andrew, for your question. And specifically with regards to the AI tailwinds that we're seeing in the rest of the business, we've talked frequently about the leverage of our communications technologies and our IP into end markets. We do have some specific technologies around semiconductor wafer test, where we're seeing significant capacity expansion there across multiple dimensions, including advanced nodes, memory, silicon photonics. As far as looking forward, the forecast around wafer fab equipment and capital going into those markets continues to expand. And so that is robust. In the general electronics space, what you're seeing is, again, to the earlier question from Meta about test intensity, you're seeing a lot of that multilayer, high-density heterogeneity that's coming through at the component level, resulting in higher test intensity on the production side, which Kailash mentioned, we see that in some of our end markets in general electronics. And again, that's moving from the board level to the component level where the tolerances are getting increasingly tighter as you look at higher frequency and higher data throughput and just expectations of high performance in small spaces. So I think that's the challenge that customers face. And the beauty of our solutions is we address those all the way from R&D into production, and so strong leverage there. On the operating margin side, yes, we're pleased with the progress that we're making there. I think historically, that business had suffered a little bit. And over the last 12 to 18 months, we've really focused on driving top line growth because that's where it always starts and then transforming pieces of the business. We're investing more in some of the software elements of the business with ESI and the Optical Solutions group. And there's been rationalization of certain pieces of the business where perhaps lower margin, and at the same time, pursuing other opportunities that we've talked about related to digital health and the grid. So I think in summary, despite the different profile and composition of the businesses within EISG today, we're confident of our ability to continue to achieve higher levels of profitability as we move forward.

I appreciate that color, Jason. That's helpful. I just had one follow-up question for Neil. In prior commentary, you discussed potentially the synergies being $100 million plus from the acquisitions. And I'm wondering now that you've completed the integration, is your estimate of those synergies potentially larger? And I'm just sort of thinking about what sort of contribution I can assume for Q4, Q1?

Yes. I would say that at this point, given where we're at, we certainly have direct line of sight to the $100 million. So I think that's -- to the extent there was risk in that, that's substantially derisked. And I think as this business now begins to operate more holistically within the Keysight framework, we'll continue to look for additional opportunities. So I don't have a quantification for you, but I think history would suggest that when the initial kind of wave of synergies come out that there is admittedly smaller dollars, but some additional follow-on efficiencies that tend to materialize. If you think about it on an incremental basis, as you move from FY '26 to FY '27, I'd be thinking on the order of $50 million. I think we realized close to $40 million, again, ramping throughout the year as you thought about it from Q1 through Q4. And again, I expect that we're going to be kind of close to 90% of that $100 million realized as we enter next fiscal year.

Your next question comes from the line of Joseph Cardoso from JPMorgan.

This is Marc Vitenzon on for Joseph Cardoso. I guess you guys have given us a lot of detail on the strength in AI-related wireline. So I wanted to ask about traditional non-AI portion of wireline. How does growth look like in that business? And what are you seeing there?

Yes. I think we're seeing a convergence, I think, Marc, is what -- how I would characterize it. And I think that is to be expected as such a disruptive technology starts to intercept multiple end markets. I'll give you an example. The wireless customer base, historically, maybe never had to think about AI, and now AI is entering the RAN. That's an example of an application. The wireless ecosystem has a known set of contract manufacturing companies that play into the telco market. Many of them have now started to invest in building their own racks for AI. That's another example of some of the applications that are now emerging that we're well positioned to capitalize given our strength and reputation in this ecosystem.

Got it. And then you guys mentioned that software and services was roughly 33% of revenue. I guess I'm curious like where do you think that number eventually goes, especially following the recent acquisitions?

Yes. I think, look, the strategy for the company is to become a solutions company. This has been what we have worked hard to building. And inherently, that implies providing more software-centric solutions and also offering differentiated services to build the life cycle value contributions. And we do it at the pace of our markets and at the pace of our customers. It's all about our customers' needs. And so we're not trying to force a business model into the marketplace. And so yes, we've trended as high as 40% a year or so ago, and now we're at 33%. But I just want to say on a dollar basis, this is record levels for software and services, and we'll continue to keep innovating to stay differentiated in the marketplace.

I would also maybe just add that our software and services businesses are also growing double digits. They're just not growing as fast as the hardware businesses at this point in time.

Your next question comes from the line of Matt Niknam from Truist Securities.

Congrats on the results. Just 2, if I could. First, on the 4Q revenue guide. So it's implied to only increase about 5% sequentially. I know that's pretty normal in terms of seasonality, but your book-to-bill has been north of 1.1 for 2 straight quarters. So I'm wondering if you can speak to any sort of supply constraints that are inhibiting or limiting that revenue guide for the fourth quarter? And if there's any color in terms of how much of the backlog that's been growing is going to ship next fiscal year? And then on a related note, I think the last several years, you've given some initial color or framework in terms of next fiscal year on the third quarter call. I'm curious if there's any initial thoughts you're ready to provide just given the momentum you're seeing across the business.

Maybe I'll take the 2027 commentary, Neil, and then you could cover the remaining. It's a great question. Look, we have no doubt a strong setup as we enter fiscal '27, even as we remain cognizant of the outperformance you're seeing in 2026 on top of the growth year in '25. Supply chain will remain the governor of near-term revenue. I think I mentioned that earlier. But when I look at the broader end market, I just have to look at the technology trends and say the complexity of these technologies are only growing. The intensity and the pace at which our customers are innovating across our end markets globally, it's relentless right now, and that intensity is matched with their investment. And we're well positioned as a company to capitalize because of all the investments we made in the downturn in R&D that is now going to generate and has already started a good refresh cycle for our new products, which are already being enthusiastically received by our customers. So we're well positioned from that point of view. We also are taking a -- given this demand that is consistently now for a few quarters, and we're executing well on a quarterly basis, but we've taken a longer-term view, 18-month-plus view of our supply chain planning, and we're working to create additional flexibility, but those come with a lag because I'll give you an example. We'll have to redesign some products on the margin to accommodate second sources and enter into some more longer-term agreements with our customers. So we're already starting to take those actions. And so I put it all together, I feel good about the setup. We'll give you more specific guidance for Q1 when we report in Q4. Neil?

Yes. I mean I think you said most of it. I think if I was just going to recap. Right now, supply is not the limiter, right? Excuse me, demand is not the limiter. We do have some supply chain limitations. It's a little bit of a mixed bag. I think if we think back 3 months to the biggest supply chain challenges we were facing, most of them were internally -- internal capacity related around ramping some NPIs that have seen kind of unprecedented early demand from the marketplace. We've made tremendous progress in that area. I think the challenges have shifted more towards incoming parts which are under high demand as demand has ramped and continues to ramp across the ecosystem, right? You've got numerous players that are all competing for supply from a similar set of component suppliers. And so I just think that the supply situation is likely to be nonlinear and will likely be a governor of our ability to convert demand into revenue for the next several quarters.

If I can just add a couple of comments about 2027 outlook. We just had the biggest refresh of our core RF microwave and high-speed digital products since we formed Keysight. And we reinforced this at our worldwide annual sales training event in June, where we trained our entire sales team on this portfolio as well as Spirent products. So as these solutions are rolled out and introduced to more and more customers, I expect this to be a strong tailwind for us for many years, many quarters to come.

And years, Steve.

Years.

Your next question comes from the line of Quinn Fredrickson from Baird.

I wanted to go back to the orders discussion. Good to see the $2 billion orders number again, but it's been unusual typically to see orders grow sequentially in the third quarter. So could you just unpack what the drivers were? And do you think you saw any tailwinds for possible future sovereign transceiver restrictions or customers getting ahead of any other constraints or supply issues?

There was nothing unusual about the demand. In fact, we saw conversion of the pipeline in a very orderly fashion, no pull-ins. We are looking for it. It's just an environment where the markets are stronger and Keysight's differentiated position in its core markets, starting within commercial comms, which really outperformed for us relative to our expectation even with AI. And equally, the demand from our prime contractors in aerospace defense remains strong as they're building out capacity. Sovereign investments in Europe is another tailwind for our defense business. And the EISG business is clearly outperforming with the semiconductor business doing exceptionally well. So strong broad-based demand, and we expect, as Steve mentioned before, we expect that to continue into Q4.

And Neil, gross margin came in at 69%. I think you had said mid-67% range. Was the difference just all incremental volume or mix? Just if you could unpack that? And then any color on how to think about sustainability into 4Q or even '27?

Yes. I mean I think if you go back a quarter and adjust for the tariff thing, we were 68% last quarter, we're 69% this quarter. It's obviously volume is helping, but I think it gets to the differentiation of the solutions that we're bringing forward across end markets, whether that's early 6G, AI, semiconductor, we have a highly differentiated set of solutions. Mix does -- we do have a broad range of gross margins across the portfolio. So mix does matter. But I do think this upper 60s percent is a sustainable level for us.

Your final question comes from the line of Bastien Faucon-Morin from SIG.

This is Bastien filling in for Mehdi. You saw some momentum in wireline. It was up 56% year-on-year. And you mentioned the mix going towards 70%, 30% production R&D from 80-20. Could you give us a breakdown on how you expect R&D and production to look like in revenues in the coming quarter?

I should say, Bastien, it's one of those things that moves -- things move around on a quarterly basis. So we tend to look at it over a longer-term horizon because in a given quarter, if a customer comes in and they're doing an expansion in production line, that could dominate a certain part of the segment. But as I've called out, I think we said 2/3 in R&D, 1/3 in manufacturing. On the margin, we're seeing in our pipeline greater activity as the customers are scaling. They're ramping production of 1.6 terabit as we speak right now, and the demand is very, very strong, right? So in a given quarter like Q4, I can easily see that mix even trend more towards production, and we're meaningfully participating across the workflow. And that's the important message is our R&D business is growing, as Kailash mentioned. We're very pleased with the diversity of that business, and we're also happy that we're participating in the volume part of the data center build-out as well.

Got it. Very helpful. And then as a follow-up, how should we think about the long-term mix of R&D versus production given that the ramp and adoption of new transceivers are accelerating? Your volume is kind of ramping, but you also have more complexity when it comes to testing those new technologies. Is there a way to think about that long-term R&D and production mix as we're reaching higher deployments of optical transceivers?

Yes. I mean I would just say, look, our strategy, as we have laid out, it's really to be an innovation accelerator for our customers. That's sort of our core purpose. We're focused on building our tech stacks, optical, electrical, both of them to be able to help our customers go through in R&D. And what we're seeing now is pretty unprecedented that the rate of adoption of these technology curves or technology waves it's accelerating to a point where you're seeing concurrent adoption across many dimensions, right? 800 gig is still sort of the underlying technology, but 1.6 is scaling. And typically, that would have been spaced out a little more. So it's really hard for us to make sense for how long this goes and how broad this goes. But I do know customers are already engaging us on 3.2 tera. We did a demo at a conference earlier this year. So the intensity associated with technology change and the economics for companies to find the latest technology remains high, and we're participating in this, and I feel good about our position heading into '27.

That concludes our question-and-answer session for today. I would like to turn the call back to Liz Morali for any closing comments.

Thank you, Hilary, and thank you all for joining us today. A replay of today's call will be available on the Investor Relations website later today, and we appreciate your interest in Keysight.

Thank you for attending. This concludes today's call. You may now disconnect.