Note: This is an earnings call transcript. Content may contain errors.

Image source: The Motley Fool.

DATE

Tuesday, Oct. 28, 2025, at 5 p.m. ET

CALL PARTICIPANTS

  • Chief Executive Officer — Scott Bibaud
  • Chief Financial Officer — Francis Laurencio
  • Vice President, Investor Relations — Mike Bishop

Need a quote from a Motley Fool analyst? Email [email protected]

RISKS

  • Scott Bibaud stated that STMicroelectronics will "take BCD110 to market without MST, and currently, they have no plan for a future variant that includes it," resulting in the loss of potential royalty revenue for Atomera (ATOM 5.05%) from this program.
  • Francis Laurencio reported a net loss of $5.6 million for Q3 2025, up from $4.6 million in Q3 2024, with operating expenses rising to $5.7 million in Q3 2025 from $4.8 million in Q3 2024.
  • The company recorded a negative gross margin, attributed to the timing mismatch between incurred wafer deposition costs and the recognition of related revenue.

TAKEAWAYS

  • GAAP Net Loss -- $5.6 million, or $0.17 per share, in the third quarter of 2025, compared to $4.6 million and $0.17 per share in the same period last year, reflecting increased costs.
  • GAAP Operating Expenses -- $5.7 million in the third quarter of 2025, an $857,000 increase from $4.8 million in the prior year period, driven mainly by higher R&D and stock compensation expenses.
  • Non-GAAP Net Loss -- $4.4 million in Q3 2025, up from $3.9 million in the same period last year, due primarily to a $423,000 rise in non-GAAP operating expenses.
  • Cash and Cash Equivalents -- $20.3 million as of September 30, 2025, down from $22 million at June 30, 2025.
  • Cash Used in Operating Activities -- $3.4 million used in operating activities during Q3 2025, slightly less than the $3.5 million used in the second quarter of 2025.
  • ATM Facility Proceeds -- Approximately $2 million was raised in Q3 2025 by selling about 393,000 shares at an average price of $5.23, with an additional $836,000 raised subsequent to Q3 2025.
  • Shares Outstanding -- 31.7 million as of the call date.
  • Q4 Revenue Guidance -- Expected NRE revenue of $75,000 to $125,000 from wafer shipments to customers, recognized over Q4 and into the next year.
  • Non-GAAP Operating Expense Forecast -- Full-year 2025 non-GAAP operating expense is anticipated to be $17.25 million to $17.50 million.
  • STMicroelectronics Platform Update -- STMicroelectronics discontinued MST integration in its BCD110 platform, confirming there is "no plan for a future variant that includes it."
  • MST Technical Validation -- A new MST implementation, validated by both Atomera and STMicroelectronics, "doubled our performance improvement without the associated reduction in device lifetime," according to Scott Bibaud, establishing a significant technical advance.
  • Customer Engagements -- Record customer wafer-processing activity this quarter, including active programs in power, Gate-All-Around, DRAM, and RF-SOI segments.
  • Commercial Priorities -- Management highlighted prioritization of MST starting wafer products and strategic OEM partnerships to accelerate time-to-revenue opportunities.
  • Executive Appointment -- Wei Na joined as VP of Sales, with prior experience scaling semiconductor technology licensing businesses targeting Atomera’s core customer base.

SUMMARY

STMicroelectronics’ decision not to incorporate MST in its latest BCD110 platform eliminates Atomera’s current path to royalty revenue from this key project, even though a new, higher-performing MST solution was technically validated by both companies. Atomera set new highs for customer wafer processing, signaling notable operational momentum across multiple application segments despite a larger net loss and negative gross margin. Management’s strategic focus now emphasizes shorter revenue cycles from MST wafer products, deeper OEM partnerships, and converting ongoing technical programs into commercial licenses within Gate-All-Around, DRAM, RF-SOI, and GaN markets.

  • Francis Laurencio explained that stock compensation expense rose to $1.3 million, up from $907,000 in the prior year period, primarily due to the adoption of performance-based RSUs that vest over three years if specified shareholder returns are met.
  • The quarter’s cash outflows were partially offset by new equity raised under the ATM facility, with share issuances continuing post-quarter close, reflecting ongoing financing efforts.
  • MST starting wafer approaches are being leveraged in RF-SOI, GaN, and potentially future DRAM products to accelerate customer qualification and time to market.
  • The technical learning and materials innovation developed during the STMicroelectronics program are being applied to other large power device customers and active engagements in additional market segments.
  • Bibaud said, "we have no lack of opportunities across several different segments," referencing ongoing work with multiple leading manufacturers in power, memory, logic, and RF device markets.
  • Commercialization timing remains variable as customers’ production integration and wafer validation cycles differ, with meaningful revenue from license agreements contingent on successful qualification and adoption.

INDUSTRY GLOSSARY

  • MST (Mears Silicon Technology): Atomera’s proprietary thin-film material designed to enhance the performance and reliability of CMOS-type semiconductor transistors.
  • RF-SOI (Radio Frequency Silicon-On-Insulator): A semiconductor substrate technology optimized for RF signal transmission and isolation in high-frequency device applications.
  • Gate-All-Around: An advanced transistor architecture in which the gate material entirely surrounds the channel region, enabling improved control in nanosheet and next-generation devices.
  • GaN (Gallium Nitride): A wide-bandgap semiconductor material used for high-performance power and RF electronics due to its efficiency and voltage handling capabilities.
  • NRE (Non-Recurring Engineering): Fee-based revenue for custom engineering services or prototype wafer delivery, distinct from recurring license or royalty income.

Full Conference Call Transcript

Scott Bibaud: Thanks a lot, Mike, and good afternoon, all. This has been a quarter of both challenge and validation, one that underscores the reality of bringing a new material technology to market and the opportunities that come when you solve fundamental problems for the semiconductor industry. I'll start by addressing our update with STMicroelectronics and depart from our regular format to review the broader picture, the momentum we're building with new customers and the different market opportunities that Atomera's technology is being used to address. As many of you have seen in our announcement, our work with STMicroelectronics on their smart power platform reached an inflection point this quarter.

During this program, we were tackling a very difficult performance trade-off for their 200-millimeter platform. We achieved what we set out to do, significant performance improvements in key device metrics. However, that higher performance came with a corresponding reduction in device lifetime, often referred to as reliability, which failed to meet all of ST's specifications. Over many months, our 2 teams worked closely to resolve this trade-off. Then ST, as part of a reshaping of its manufacturing footprint, announced they would discontinue development on 200-millimeter wafers to focus exclusively on 300-millimeter for the next-generation BCD110 platform.

At about the same time, Atomera discovered a new MST implementation validated through our TCAD simulations that doubled our performance improvement without the associated reduction in device lifetime. In other words, we found a way around the trade-off, and improvement only made possible by using MST. Over the last few months, ST validated our findings for the new implementation. However, because this new version required a device architecture change that would take multiple learning cycles to validate, they determined that they could not incorporate it and still meet their aggressive BCD110 launch schedule. Therefore, ST informed us that they will take BCD110 to market without MST, and currently, they have no plan for a future variant that includes it.

That means we no longer have a line of sight to royalty revenue at ST for this particular program. While that outcome is certainly disappointing, there are several important positives I want to emphasize. First, at STMicro, we demonstrated significant performance gains and proved MST's integration capability inside a Tier 1 production fab. Second, we've now developed a very high-performance solution that eliminates the performance reliability trade-off, which is a significant new differentiator for us going forward, one that we are already actively discussing with other players in this market. And third, ST has reiterated their intent to continue working with us in other technology areas where MST could add value.

Under their license with us, they continue to run experiments across several different businesses. This chapter with ST underscores that moving a new material into mass production is rarely linear. But the learning from this effort makes us a stronger -- gives us a stronger foundation as we engage with others in the same power market segment, including with a very large existing customer and even a new engagement that began this past quarter. Customers are now evaluating MST for power devices between 5 volts and 48 volts. It's important to keep in perspective that ST is only one of many large customers we are working with today to take MST into production in the power area.

We also have 3 other very active technology focus areas. In the Gate-All-Around space, there are 3 large competitors and one that's still emerging. We're working with or in discussions with all 3 of them. I mean, all 4 of them. In the DRAM space, there are 3 large manufacturers, and we are engaged with 2 of them right now and have good relationship with the third. In the RF-SOI area, we're doing integration work with 4 different fabs and a fabless player right now with many of them running wafers. So you can see that we have no lack of opportunities across several different segments.

Indeed, during the last 3 months, we processed a record number of wafers for our customers. When we look at all these opportunities, it's helpful to understand how we prioritize our business in terms of revenue potential. The first being the fastest time to market, second being the highest return on investment and the third being breakthrough long-term growth. One of the fastest ways to get Atomera's technology to market is through applications which use MST deposited on top of the starting wafer rather than inserted into the middle of the manufacturing line. There are many reasons why this can accelerate revenue.

First, customers can simply acquire an MST starting wafer and run it through their standard production flow with very few process modifications for an easy experiment. They don't have to install MST, deal with the complications of wafers being transferred in and out of their fab, make major changes to their process to integrate it or complete a license agreement. The price of MST can be built into the cost of the starting wafer, which gives Atomera the same revenue, but the customer will not view the cost as a royalty. And it's certainly faster to get MST starting wafers qualified than something integrated into the middle of the process.

Today, we use MST starting wafers in our work in RF-SOI, in GaN and possibly soon in next-generation DRAM. We actively seek out these implementations because of the relatively easier integration and shorter path to revenue. The second set of applications have enormous revenue potential, but the development process can be more demanding because MST is inserted into the middle of a complex set of production steps. It is worth it, though, because the upside represents a massive return on investment, including in the areas of Gate-All-Around logic, DRAM, power devices and other memory products. One design win here will ensure the future success of the company.

And as I mentioned earlier, we have at least 6 or 7 of those efforts underway today. In Gate-All-Around and advanced memory, our partnership with a leading capital equipment company announced earlier this year is showcasing our competence at advanced nodes. Using their test infrastructure, we've been able to validate MST's ability to reduce contact resistance, improve channel reliability and be deposited in the tiny structures of nanosheet transistors. We are very excited by the deep cooperation and customer interest generated through this partnership. This quarter, we'll be hitting the road on joint visits with our customers to persuade them that issues in their manufacturing process can be solved using MST. The weight of our partners' endorsement cannot be overstated.

Finally, we have an abundance of new breakthrough materials enabled by MST under development in the background through commercial partnerships and university collaborations. For many of them, we've already filed fundamental patents, and we're now in the process of making prototypes and understanding their capabilities. This is the type of program, for instance, which launched our GaN work. We have a dozen similar initiatives in early investigation, several of which might become near-term disruptive technology announcements in areas like quantum computing, AI server power, high-bandwidth memory architectures, piezoelectric devices, optical networking and a variety of other areas, which have the potential to enable entirely new applications.

Farming out the early R&D whenever possible, allows Atomera's core team to keep a laser focus on the nearer-term revenue opportunities and apply more resources only when we see the potential of these innovations coming to fruition. Our gallium nitride initiative continues to deliver exciting progress. In collaboration with Sandia National Labs, we're in the process of completing device fabrication to highlight our improved electrical performance. Prior results have confirmed MST's ability to enhance GaN growth on silicon substrates, a major barrier for high-volume production and have garnered interest from our first commercial customers. We hope to release a complete data set publicly later this year, which will be the precursor to a full-scale rollout.

As we continue our GaN work with Sandia, they are now seeking to expand the areas of R&D engagement on a range of Atomera technologies corresponding to their highest priority development areas. The semiconductor industry is clearly entering a new materials innovation cycle. Across logic, memory, power and RF, engineers are hitting the limits of conventional scaling. They're searching for material solutions that can boost performance, improve reliability and reduce variability, exactly where MST delivers value. This is particularly true in AI infrastructure and data centers, where the demand for power efficiency and thermal management is driving renewed focus on device-level innovation, which MST can deliver.

One of our principal challenges is to ensure that potential customers know about MST -- and that is why I'm so excited to welcome Wei Na as our new VP of Sales. Wei has had experience growing a semiconductor technology licensing business very much like Atomera from scratch, selling to the exact same customers we are addressing, and we believe his leadership will help us both grow sales and convert existing opportunities into licenses.

Our priorities remain clear: emphasize MSD starting wafer products like RF-SOI and existing engagements to get to production and revenue as quickly as possible; 2, leverage our strategic OEM partnership to advance active engagements in Gate-All-Around logic, memory and power through our comprehensive silicon test results and early licenses; 3, bring MST for GaN technology to a customer-ready stage with shareable electrical data; and 4, maintain fiscal discipline as we transition from R&D validation and integration to revenue-generating licenses. Our mission hasn't changed. It's to enable better, faster and more efficient semiconductors through advanced materials engineering. That mission remains as relevant as ever. I want to thank our employees, our customers and our shareholders for their continued confidence and support.

Every quarter, we move closer to the point where MST's impact will be felt across multiple product lines and foundries worldwide. With that, I'll turn the call over to our CFO, Frank Laurencio, to review our financials.

Francis Laurencio: Thanks, Scott. At the close of the market today, we issued a press release announcing our results for the third quarter of 2025. Our summary financials are shown on this slide. Our GAAP net loss for the third quarter of 2025 was $5.6 million or $0.17 per share compared to a net loss of $4.6 million, which was also $0.17 per share in Q3 of last year. GAAP operating expenses in the third quarter of this year were $5.7 million, an increase of $857,000 from $4.8 million in Q3 of 2024.

This was due to a $544,000 increase in R&D expenses, reflecting both higher outsourced device fabrication work and increased compensation expenses and the $353,000 increase in G&A expenses, primarily consisting of higher stock compensation expense. Sales and marketing expenses were basically flat. Non-GAAP net loss in Q3 2025 was $4.4 million compared to a loss of $3.9 million in Q3 of last year due to a $423,000 increase in non-GAAP operating expense, primarily reflecting the higher R&D expenses I just discussed. Stock compensation expense, which is the main difference between GAAP and non-GAAP operating expenses was $1.3 million in Q3 of 2025 and $907,000 in Q3 2024.

The increase in stock compensation expense, which is noncash, reflects the adoption of performance-based RSUs or PSUs for executive equity-based compensation in March of last year. PSUs vest over 3 years rather than 4 years as is the case for time-based RSUs. However, PSUs will only vest if we deliver shareholder returns that meet minimum targets relative to the Russell 2000 Index. Sequentially, Q3 2025 non-GAAP net loss of $4.4 million compares to a $4 million net loss in Q2, primarily due to higher R&D expenses. Our balance of cash and cash equivalents as of September 30, 2025, was $20.3 million compared to $22 million as of June 30, 2025.

We used $3.4 million of cash in operating activities during Q3 compared to $3.5 million in the second quarter of this year. During Q2 -- sorry, during Q3, we raised approximately $2 million under our ATM facility, net of commissions and expenses by selling approximately 393,000 shares at an average price of $5.23. Since the end of the quarter, we've raised an additional $836,000 from sales of approximately 171,000 shares at an average price of $5.03. As of today's date, we have 31.7 million shares outstanding. In Q4, we expect to recognize between $75,000 and $125,000 of NRE revenue from wafer shipments to customers running the demos that Scott mentioned in his remarks.

Those shipments and the associated revenue recognition will happen in Q4 as well as into next year. Gross margin was negative this quarter because a portion of the cost for MST deposition on those wafers was incurred during this quarter, but the revenue will be recognized as we ship the wafers going forward. Moving to expenses. I expect our non-GAAP operating expense for the full year 2025 to be in the range of $17.25 million to $17.50 million. Sales and marketing expenses ticked up last quarter in connection with recruiting for both sales and marketing leadership roles. The compensation expenses associated with those roles are built into our plan.

Our recruiting efforts have started to pay off with the hiring of Wei Na as our VP of Sales. With that, I'll turn the call back over to Scott for a few summary remarks before we open the call up to questions. Scott?

Scott Bibaud: Sorry, a little trouble with the Zoom controls here. Thanks, Frank. Across all of our technology focus areas, we have strong developments underway with the leaders of the industry. I hope today, we've given you a sense of our wide and deep potential to deliver important material solutions that will ultimately make Atomera a financially successful technology provider across many different semiconductor segments. I appreciate you taking the journey with us. Mike, we will now take questions.

Mike Bishop: All right. Thank you, Scott. [Operator Instructions] Right now, our first question comes from Richard Shannon of Craig-Hallum.

Richard Shannon: All right. Great. Hopefully, I'm unmuted here, Mike.

Mike Bishop: You got it.

Richard Shannon: All right. Excellent. Thanks Scott and Frank, let me ask a few questions here. Scott, maybe let's do a redux on STMicro. So I guess my first question here is, so it sounds like you did a new design on 300 millimeters that you validated in your simulations, but there would have been multiple cycles of learning to validate for ST. So is that trying to match your simulation to the real world to their simulations to make sure that it worked, and that cycle time was just too much to fit within their time frame getting to 300-millimeter. Is that the kind of the dynamic here that led them to their decision?

Scott Bibaud: Yes. So first of all, the work -- the new implementation we came up with would have worked on 200-millimeter or 300-millimeter. And actually, if you let me digress one second here, Richard, because we've gotten a number of questions that have come in where people were asking, when did you know about this trade-off between the reliability and performance. Every time you do a development, it's about trade-offs. You're doing a trade-off on one thing -- I mean, you get -- that's why we always talk about cycles of learning. You get some big improvement in one area, it breaks something else.

And then you have to go in and you have to work to fix the other thing and try to get to a point where it's all balanced out. So this trade-off work that we were doing is not at all unusual. It's what we do with every customer all the time.

What is unusual is that because they made the transition from 200 to 300, (sic) [ 200-millimeter to 300-millimeter ] we lost the ability to bring in that ultimate solution and get it done for them in time because the 300-millimeter delayed their development efforts and then they needed to get into production fast, and so they just didn't have time to run the validation runs to get our new thing proven out. I'm not sure that answered your whole question. Let me know.

Richard Shannon: I guess the point here is that it sounded like they were confident that this solves not only the performance, but the reliability issue that you discovered in 200-millimeter, and it was just the time frame that was too tight for them to want to continue right now?

Scott Bibaud: Yes, that's right. Originally, you asked about the simulation work. So we do simulation based on what we believe a customer's process -- manufacturing process is, but that's usually very secretive. They don't give anybody that information exactly. We can make our best approximation. And so we made a TCAD simulation that showed, yes, we really got this great improvement. And we gave it to them in this summer. And then they spent the next 2 months running their own simulations. Their simulations are very exact to their own manufacturing process. And so what they did was they put in all the improvements we saw -- we proposed.

And they came back and they said, "You know what, when we run our simulation, it also brings that level of improvement." So ultimately, I mean, the good news here is that they confirmed it. It makes us feel very confident to bring it out into the market as a new product. And it also makes us confident that at some point in time, we're hopeful we can reengage with ST on that particular product and have them take it forward and make it -- put it into their process.

Richard Shannon: Okay. All right. Fair enough. Let me follow up on one other comments you made related to STMicro and then we'll move on to some other topics here. So what seems obvious and you just commented on is the ability to take some of the learnings from the process with ST and take it to other customers in the power space here. What have you been able to do so far? Can you use similar kind of structures that you've built with ST and use those with other power customers? Maybe just kind of give us a sense of the benefits you can see from the situation.

Scott Bibaud: Yes. Exactly. So what we did with ST, there's a technique and architecture that the industry has known about for some time, but it hasn't been implementable. When someone builds it, it causes too many things to break and nobody has ever been able to get it to work. But because of the way MST works, because of the way it prevents dopants from diffusing uncontrollably, we believed that we could get that process to work. So this is not like something no one's ever heard of. It's something that -- one of those theoretical things that no one has been able to get working well and now we can get it to work well.

And so yes, it's -- we're not taking anything from -- any proprietary ST information. This is like a standard design technique that we can suddenly make work because of MST. And so yes, we can take that out to other customers, and they kind of understand the concept immediately.

Richard Shannon: Okay. All right. Fair enough. Let's move on here. In the last number of quarters, you've talked about transformative customers here. And unless I missed something, you didn't necessarily use that phrase here today in your prepared remarks. But I think you did mention a large demo run, which I think refers to one of them. And I think is also contributing to some of the revenues this year. I will ask a question to Frank on the revenue side here in a second here. But maybe just kind of detail where we're sitting with the transformative customers. And I do want to hit on one specific point that I had a question on.

I actually asked Mike Bishop offline earlier today, and he said to ask this question of you, which is you've talked about 2 or maybe 3 of these customers. I want to make sure how many we're talking about and which ones are still ongoing versus any ones that may be stalled. So if you can enumerate that first and then discuss what's going on with this large demonstration you talked about last quarter, and I think you briefly mentioned today, that would be great.

Scott Bibaud: Okay. Yes. I know everybody is frustrated with the code words, and I am too. But we -- so in January or February, we unfortunately had to announce that one customer that we had called transformative had discontinued our -- we were negotiating a deal and they had backed out of the deal. And that customer, we continue to have good relations with them. We talk with them regularly, but we are not on an active engagement with that customer right now. In that same call, which I think was early -- was in February. We mentioned 2 new transformative customers that were getting underway. And yes, we are working very actively with them.

When we talked about a record number of wafers that we're processing, it includes those 2 customers that we call transformative back then. And so now today, I mentioned these 4 different segments and how we're working with a lot of customers. And then I broke it down by revenue potential and the folks in the middle, folks that are doing Gate-All-Around, folks that are doing DRAM, there are really big players who are doing power and other memory architectures. They are all massive and they're all customers that I would call transformative and so we are -- we're working with more than just those 2 that I mentioned on the call.

Richard Shannon: More than just the 2 that you would refer to as transformative. Is that what you're saying, Scott?

Scott Bibaud: Yes, yes.

Richard Shannon: Okay.

Scott Bibaud: I mean I -- just discontinue the term transformative. [ We're good. ] These 2 customers I spoke about as transformative in February are just very, very large revenue potential customers with very big processes that we hope to get going on. But we're also working with other customers who are also very large and have the potential to be transformative.

Richard Shannon: Okay. Well, let's talk about the specific transformative customer you talked about last quarter that you're doing a large demo run here. What's the update on what's going there? And is that leading to at least some contribution to the revenues you're guiding to this quarter?

Scott Bibaud: Yes. Maybe I'll let Frank answer that. But -- so one of the -- there's some trickiness about when we book revenues. And so we have a lot of customers. The revenue that we're putting out this quarter is based on several customers. I can't answer whether that specific one is in Q3 or it will be in the guidance that Frank gave for Q4, but it's -- yes, we're getting revenue from wafer runs with that customer.

Francis Laurencio: Yes, that's right. I mean, the revenue guidance actually covers multiple customers, 3 different customers. And it's spaced out over time. And while I don't like to show negative gross margin, the timing issue gives a little bit more visibility in the sense that we do a bunch of the deposition work, which is when we incur the cost of our tools, the metrology and the labor associated with it. And oftentimes, these are -- these can get matched up pretty quickly with the revenue because it's a small number of wafer runs, and that's been true in the past.

But we've been talking now for a couple of calls that we've been working with a very large customer on one of the largest wafer -- on the largest wafer run that we've ever done. And we also have other customers. So now what you're seeing is, we do a lot of that work we don't ship all of those wafers out, but we don't necessarily do all the deposition because the nature of these engagements is it can be iterative. You may do some wafers for setup, you run a series of tests, the customer validates those, you get some feedback.

You then do another run with slightly different conditions, either on the MST or how the customer processes it with implants and things of that nature. So you can get a lot of activity in 1 quarter and then the wafers will ship out over time. And one of the challenges in sort of giving guidance is, it isn't set in a schedule of we're going to ship 25 wafers this month and 25 wafers 2 months after that. Sometimes it really depends on what the customer learns in the process of evaluating that, setting up a new set of experiments and then we ship out more.

So yes, there's multiple customers here, and these are important engagements in different application areas.

Richard Shannon: Okay. All right. That's helpful, Frank. I'll probably follow up with you a little later on that one. Maybe 2 more questions. I will jump out of line here. First of all, Scott, in your prepared remarks here, and I'm sure we'll review these in detail when the transcript comes out here, but you talked about kind of segmenting your opportunity based on where in the stack your MST is applied here and you talked about on top of the wafer versus somewhere in the middle. Certainly, layers in the middle or -- I think it's fairly understood, especially for me who is not a device guy per se that's very complicated.

But vice versa, if you can apply just on the top, that seems to be a much simpler process, which also implies it might be an area where by which you might expect to see or hope to see your first license here just from a time-to-market perspective. So 2 questions for you is, I think I missed the applications areas that, that specifically applied for. And b, would you agree that, that's a very -- a somewhat likely or very likely situation by which you first reach first manufacturing license and commercial production?

Scott Bibaud: Yes. So first of all, yes, you're right about being deposited on top of the wafer makes it much easier. The applications that we specifically spoke about that do that is RF-SOI and gallium nitride; and also in the future, we have some ideas on next-generation DRAM that could use it. So one thing to understand very briefly is when we deposit MST in the bottom layer, it has to be on a process that doesn't use incredibly high heat for long periods of time.

So if we deposited -- if it was on an MST starting wafer and then someone put the wafer into an annealing step that was 1,100 degrees for an hour, then that would really damage the MST itself and it wouldn't work. So the only time we use MST on the starting -- on the start of a wafer is on -- manufacturing processes are going to be lower temperature. And there's a lot of those. Like RF-SOI is running at very low temperatures. The new Gate-All-Around processes, they're trying to run them at very low temperatures. So in theory, MST could be on the base -- on the starting wafer for those.

Gallium nitride, we put MST on bottom before it grows the gallium -- yes, the gallium nitride on top of it. That one isn't quite as low temperature, but it doesn't matter. The MST still works as a starting wafer. So I think a layman might say, well, why don't you just do every process as a starting wafer if it's much easier and faster time to revenue. Well, it has to fit a certain dynamic, which has to do with this temperature range. You had a second half to your question, and I've talked those [ I might have ] forgotten it.

Richard Shannon: You hit the applications. I think you've answered most of it. So I think that's very helpful. Last question for me, I'll jump out of line. You talked about this large capital equipment partner. And I think today, you mentioned about going on a roadshow here. Maybe just kind of give us a sense of how broad the engagements are with this company. I think in the past, you mentioned 2. I don't know if that was the limit or there were more you just didn't mention, but -- well, how do we understand the scope and breadth of your interaction with customers through or with them?

Scott Bibaud: Okay. So the stated aim of our partnership is in the Gate-All-Around market. And that was what we announced in our press release. However, I have to say that -- there's great value in this partner working with us in everything. And there's value in us working with them in everything. So we have talked to them a lot and done some work on DRAM as well. So basically, yes, I would say our primary focus right now is Gate-All-Around and DRAM. And when we go out on the road, that's who we'll be really targeting most closely.

Mike Bishop: Okay. Thank you, Richard. A number of questions have come in on the Q&A line, and I will aggregate them and ask some of the more common ones. So first one is about the Gate-All-Around projects and when the -- there's a number of current projects underway that are expected to launch soon. And how many years do you expect the target process you are currently collaborating on to enter production?

Scott Bibaud: Yes. So first of all, working with a few different customers, so there might be a different answer for each customer. In general, the guys working on Gate-All-Around, the great news is it's amazing working with them because they have armies of people working on this stuff, lots and lots of resources to test out your material. And the bad news on that is that they come back with a ton of requests for more information and more testing. But they're almost always working towards some kind of a launch that you would be built into. Some of them, I would say the majority are looking at a launch that's still a few years out.

There is some of them that are actually looking at using MST to improve yield on processes that are in production today. I can't exactly say, well, if or how long it would take to get into production on those processes. But my guess is if they integrated MST, they would have to do some qualification work on it. But if it did indeed improve their yield, which I think is what the majority of them are looking at for the current timing processes, they would try to move it into production very quickly.

As long as it didn't break anything in the specifications of their production wafers, they would have every incentive to get it into production as soon as possible to improve yield.

Mike Bishop: All right. In the past, you've talked about JDA1 and the fabless RF licensee. Have you been doing wafer runs for those? And what do those results look like?

Scott Bibaud: Yes. So the answer is yes, we are doing wafer runs with them. Unfortunately, we don't have the results yet. I can't really commit that I'll be able to give you results from each customer. But generally, what happens is when the results come out, that's the timing when we'll be able to start driving towards licenses and transitions to production. Generally speaking, we have a number of different customers with wafers underway right now. None of them are coming out in the next few months. I would say we might have some coming out at the end of the year, but more likely into the first quarter before we start seeing a lot of results from those runs.

Mike Bishop: Okay. And one for Frank. So the Incize partnership for GaN testing, can you talk about the economics there of who's paying for the runs and -- or for the testing, if you could shed a little light on that?

Francis Laurencio: Yes. I mean at this stage, this is a -- an arrangement with [ RF ] Incize, where we're each bearing our own costs and we'll hopefully achieve a result that would lead us to some further activity. But right now, it's -- we're not paying them to run testing nor are they paying us for wafers. So it's early stage. And I think our hope right now would be to generate good RF data because that's something notoriously difficult. RF testing is complex. It's not something that we can typically do ourselves. So a lot of the work on RF-SOI that we can do is kind of physical characteristics of our film.

But when you get into some of the testing of actual devices on kind of different figures of merit, then those are more specialized tests. And so getting more insight into that is very helpful from a marketing standpoint. And our view is there was some question on work with Soitec and wafer-based products. The more information that we have to market to the ultimate customers of RF-SOI devices, the better it is in terms of building a relationship with Soitec, who's a wafer manufacturer. So the more end demand that they see, the closer the collaboration is with us. So I kind of see it as a means to an end there.

Mike Bishop: Okay. And then Scott, going back to a topic we've touched on in the past, but is there an update on JDA2?

Scott Bibaud: JDA2 is running wafers with us. And they're one of the ones that I talked about that we'd hope to get some results at the beginning of the year; and hopefully, see if we can turn that into a license and then plan to go to production.

Mike Bishop: Okay. And then with regard to the STM news, we had a number of questions on disclosure channel. And can you talk about why you chose to put the news out on a blog post?

Scott Bibaud: Yes. Yes. And we went back and forth on that. So I just want to be clear, we were in discussions with ST all through August, September and into October about implementing this new version -- a new architecture we had and moving forward on 300-millimeter, and we were waiting to find out from them what the plan was, when that work would start, when they had planned that it would be trying to take it to production. And it was really just 1.5 weeks ago that we had a call with them, and that's when they told us that they did not have a plan in place to use MST to do that new architecture.

So immediately after that call, we got off the phone and we started talking about, okay, we have an earnings call in 1.5 weeks. but it seems too long to wait for 1.5 weeks before we notify investors. And so on the following Monday, we actually started speaking with ST to make sure that when we disclose this, we would be following their internal guidelines on what we could say and couldn't say. And then on Tuesday, we put out the blog post. We could have put out a press release, but press releases tend to be, at least in our opinion, much more black and white about news that you're giving.

In this case, we see it as a much more nuanced message. ST was telling us we're not -- we don't have a plan to use you guys on this next run. Yes, very bad news because I know all the investors want to know when the royalties will start flowing, and so do we. But they didn't say they'll never use us. And they also reassured us again and again that they are continuing work using our technology on other process areas. So we felt that using a blog would allow us to give a little more nuance than a press release.

And we know that the channels of communication that we have with the blog, we push it immediately out to all of our investors, so -- that are at least registered with us. And so we felt it was a good channel of communication in this particular case. And the most important thing to us was to get it out there as soon as we can within the restrictions of making sure we were working everything out with ST and so forth.

Mike Bishop: All right. And one more question here. Is there any chance of government funding now that Atomera has been working with Sandia for a while?

Scott Bibaud: I talked a little bit on this call, which I've never done much about in the past about all of the different R&D efforts that we have underway. And many of them are, as I mentioned, through Academia, through outside commercial partners so that we don't have to burden our internal team with too much of it. But Sandia is very interested in many of those technologies, and they have government programs that are interested in implementing things that would use those. So yes, there's a lot of interest through Sandia.

And we also continue to work with the government and with the CHIPS Act infrastructure such as it is to see what we can do to kind of deliver some of our technology in through that channel and get some near-term revenue that way as well.

Mike Bishop: Okay. Thank you, Scott. At this time, we'll turn the call to Scott for closing comments.

Scott Bibaud: Okay. Thanks, Mike. Okay. Yes, thanks for joining us and listening to our progress that we've been making here at Atomera. Next month, we'll be attending the Craig-Hallum Alpha Select Conference in New York, and we look forward to seeing some of you there, if you'll also be attending. Please continue to look for our news articles and blog posts, which are available along with investor alerts on our website, atomera.com. Should you have additional questions, please contact Mike Bishop, who will be happy to follow up. Thanks again for your support, and we look forward to our next update call.

Mike Bishop: Thank you. This concludes the conference call.