Q2 2025 Earnings Call Transcript

Thank you for standing by. My name is Greg, and I will be your conference operator today. At this time, I would like to welcome everyone to today's Bloom Energy's Second Quarter 2025 Financial Results Call. [Operator Instructions] I would now like to turn the call over to Michael Tierney, Head of Investor Relations. Michael?

Thank you, and good afternoon, everybody. Thank you for joining us for Bloom Energy's Second Quarter 2025 Earnings Call. To supplement this conference call, we furnished our second quarter 2025 earnings press release with the SEC on Form 8-K and have posted it along with supplemental financial information that we will reference throughout this call to our Investor Relations website. During this conference call, both in our prepared remarks and in answers to your questions, we may make forward-looking statements that represent our expectations regarding future events and our future financial performance.

These include statements about the company's business results, products, new markets, strategy, financial position, liquidity and full year outlook for 2025. These statements are predictions based upon our expectations, estimates and assumptions. However, as these statements deal with future events, they are subject to numerous known and unknown risks and uncertainties as discussed in detail in our documents filed with the SEC, including our most recently filed Forms 10-K and 10-Q. We assume no obligation to revise any forward-looking statement made on today's call. During this call and in our second quarter 2025 earnings press release, we refer to GAAP and non-GAAP financial measures.

The non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between the GAAP and non-GAAP financial measures is included in our second quarter 2025 earnings press release available on our Investor Relations website. Joining me on the call today are K.R. Sridhar, Founder, Chairman and Chief Executive Officer; and Maciej Kurzymski, our Acting Principal Financial Officer.

K.R. will begin with an overview of our progress, and then Maciej will review financial highlights for the quarter. After our prepared remarks, we will have time to take your questions. I will now turn the call over to K.R.

Good afternoon, and thank you for joining us today. Bloom had an excellent quarter, the highest revenue and most profitable second quarter in our 24-year history. When the company was founded and again, in our IPO prospectus 7 years ago, we painted a bold vision to become the power provider of choice for the digital world. Over the last couple of calls, I've told you that our business is at an inflection point as demand for clean, reliable and rapidly deployable power is surging.

Now there is tangible evidence. Six months ago, we announced a strategic partnership with a major U.S. utility company, American Electric Power. Yesterday, AEP announced that Amazon Web Services and Coralogix, both data center operators, are deploying Bloom systems in Ohio. AEP's CEO, Bill Fehrman, noted that demand for power is, "Growing at a pace I haven't seen in my 45-year career." But he notes interconnection agreements take 5 to 7 years in many U.S. states, even for AEP, the largest owner of electric transmission systems in the U.S.

To avoid such a long delay, Bill also added that AEP is giving its customers solutions so they can come online quicker. Fuel cells will get AWS and Coralogix up and running quickly. Indeed, AI companies need power at AI speed, waiting 5 to 7 years is untenable and Bloom moves at AI speed. Just last week, for instance, we announced our partnership with Oracle to power their AI data centers.

We have committed to having power available to their first data center in 90 days. Time to power is one of many value propositions Bloom brings. We are also cleaner, more reliable and more cost effective than alternatives. Because our power systems are designed and purpose-built for data centers and other mission-critical applications, our installations do not require the band-aids that turbines and engines need to power data centers.

For example, we don't need multiple AC to DC converters or specialized equipment to suppress harmonics. Eliminating these band-aids enable data centers to lower costs, increase reliability and reduce carbon footprint. We are excited to collaborate directly with Oracle to help them leverage all of the benefits the Bloom platform provides. The result, Oracle can optimize the watts-to-flops ratio, resulting in increased revenue growth and margins.

Commercial and industrial customers are also increasingly valuing the velocity with which we operate. Quanta Computer, for instance, builds the AI servers that are used in the AI data centers. Their demand growth is highly correlated with AI data center demand growth. We informed you about our rapid deployment at their Fremont facility last year.

Happy with our execution, they ordered an islanded load following microgrid, which we installed in Q2. We expect new orders from other AI hardware ecosystem players soon, complementing demand we see from our more traditional commercial and industrial customers. Bloom is clearly delivering for customers. And our strong fundamentals mean we are also delivering for you, our investors. This quarter, we had record profits and operating margin, the third quarter in a row that we are hitting similar marks.

Our service business has been profitable for 6 quarters in a row. And for the first time ever, we had double-digit percentage margins, evidence of our increased reliability. And we refinanced our debt notes that were previously due in 2025, providing increased optionality to finance growth. One other highlight for the quarter, U.S. lawmakers and the administration restored tax credit benefits to companies who install our fuel cell systems.

Our country's leaders recognize that baseload power is critical to winning the AI race, reshoring factories, creating jobs and growing the economy. The tax credit will be another tailwind as we continue to grow our business. Let me close by reflecting on our position. Bloom is in a strong place.

In the 12 years since we began shipping product, we have generated over 40 terawatt hours of electricity. We have deployed more than 22,000 energy servers, our power generators, totaling well over 1 million fuel cell stacks. Each of those 1 million-plus fuel cell stacks has a unique digital twin. And over our history, we have collected over 4.5 trillion data points from the field.

Now thanks to AI, we are unlocking new ways to improve our performance, reduce costs and deliver more value to our customers. We are operating at scale and are scaling with purpose. Now our robust product has robust demand. We will double our factory capacity from 1 gigawatt a year now to 2 gigawatts a year by the end of next year.

Our mission has never felt more urgent, and we are ready. I'll turn it over to Maciej now, and I look forward to answering your questions.

Thank you, K.R., and good afternoon, everyone. As K.R. mentioned, selection of our fuel cell energy service by Oracle to power their cloud computing is another proof point for how well our technology is suited for on-site, highly reliable and variable load following power. I am thrilled to see further adoption of our fuel cell technology by leaders in the AI space. Beyond that, I'm going to limit my comments to our Q2 financial performance.

A consistent theme at Bloom has been a relentless focus on our product cost reduction and discipline around all other spend to drive profitable growth. The first half of fiscal 2025 is evidence of those efforts. Equally important, commercial execution was strong. All of this yielded a strong second quarter and first half for us.

Highlights included record second quarter revenue and gross margin and our sixth consecutive quarter of profitability in our service business. As a reminder, I will focus my discussion on non-GAAP adjusted cost and profitability metric. For a reconciliation of GAAP to non-GAAP, please see our press release and the supplemental deck on our website. Revenue for the quarter was $401 million, up 19.5% year-over-year.

Gross margin was 28.2%, 650 basis points higher than the 21.8% gross margin in Q2 of 2024, attributable to mix and level loaded manufacturing. As we said last quarter, we took advantage of our balance sheet and our visibility into customer demand to maximize efficiency and level load our factory during the first half of the year. We expect to work down this inventory as our shipment of product accelerate in the second half of fiscal 2025. Our operating income was $28.6 million versus $3.2 million loss in Q2 last year. Adjusted EBITDA was $41.2 million versus $10.2 million in Q2 of 2024, while EPS was a positive $0.10 versus a loss of $0.06 a year ago.

Again, these are all non-GAAP results. With this quarter, we have now had a profitable service business for 6 consecutive quarters. We expect this trend to continue together with margin improvement. Finally, during the second quarter, we refinanced $113 million of our convertible note that was due in August 2025 to provide more optionality to fund future growth.

It was exchanged into our existing 2029 convertible notes. Turning to the full year. We are reiterating our 2025 guidance. As a reminder, we expect 2025 revenue of $1.65 billion to $1.85 billion, non-GAAP gross margin of approximately 29% and non-GAAP operating income of $135 million to $165 million.

We expect positive cash flow from operations around the same level that you saw in fiscal 2024. We also expect CapEx to be around the same level as fiscal 2024. As we have mentioned before, we expect to see similar revenue seasonality with roughly a 40-60 first half, second half split. We are committed to maintaining strong fiscal discipline as we continue to scale. To conclude, we delivered record Q2 financial results and are reiterating our 2025 guidance. Our fuel cell solution was built for this moment when on-site, scalable, reliable, low following power is required in a matter of months.

We are well-positioned to meet the moment. Operator, we are now happy to take questions.

[Operator Instructions] And it looks like our first question comes from the line of David Arcaro with Morgan Stanley.

I'm wondering if you could elaborate on your recent success with hyperscalers. How are you seeing Bloom servers being used? Are they the exclusive power source for these AI data centers? Are these large-scale deployments? And maybe broadly with that customer set, could you see this Oracle partnership potentially act as an accelerant and spur more additional deals?

David, first, welcome to the family of analysts that are covering us. Nice to have you on. And yes, look, the Oracle deal, is the first time we as a company are directly interacting with the hyperscaler as our customer. And here, yes, it is a -- it falls in the size class of the AI data centers and it will be one single data center that the first project will power and we are working with them on many of the projects.

And here again, it's an islanded power. It is not connected to the grid. And so we carry the primary and the secondary load for this particular customer. So it is significant.

It is exactly what we have built this -- our architecture for. They will be able to use it to a lot of capability Bloom is able to offer, and it offers an opportunity for us to get started there and keep improving the number of capable attributes that Bloom brings flowing all the way into the data center and thereby optimizing both capital cost and operating cost for the customer. So we see this as extremely significant, and we are the primary source, and it is load following. So it will prove that we can load follow at large scale.

It will prove that we can operate at large scale and most importantly, AI speed. It will prove that we can install stamp sizes at that level within the 90 days that we have told you we would do in our opening remarks.

Excellent. Certainly quite an accomplishment. And maybe following on that, what gives you the confidence here now to double your production capacity? Are you -- do you also have kind of an inflection in visibility? And does the backlog give you confidence there? Or is it the underlying fundamentals of the industry, which have certainly been indicating strong demand here?

Thank you, David. Yes, we have always stated when we've been asked that we will never get past our headlights. So we have told you in the past that in the last 2 quarters, we have seen strong commercial activity. We have told you that it is very diverse, and it is high quality.

At this point in time, when we look at that pipeline, it has gotten us to a level of confidence where we absolutely feel like this is the right thing to do. That's why we are expanding the capacity, number one, right? Number two, this should be fairly simple, and it should be mind-boggling for all of us, and we shouldn't get numb to this fact. The large hyperscalers put together are going to spend more than $1 billion a day on CapEx, weekday and weekend.

It's more than $500 billion are going to be spent just in this calendar year by those people. So you take that number of $500 billion and you say, an order of magnitude down, at least $50 billion of power capital equipment needs to be spent to electrify that additional demand that's going to come on. And you take that and you do a simple math and say more than one sizable nuclear power plant's worth of baseload is what is needed every month. And we all know what can be done by the existing legacy electric infrastructure in this country.

It cannot move at AI speed. We are an obvious solution. It is self-evident to me that our demand is going to be high. And it would be wrong for us not to go and invest that money because we have that level of confidence, not just from what we are seeing in the pipeline, but when we look at this kind of investment, and we look at where is the tail for this.

This is a secular trend. It is not a 1-year or a 2-year trend. You put that together, absolutely, we should be building these factories. And the 2 gigawatt is a start to multi-gigawatts we will keep building over time, absolute confidence.

And our next question comes from the line of Mark Strouse with JPMorgan.

This is Michael on for Mark. I guess just a follow-up on that last question. How long do you expect it to take to build out this capacity? And I guess, what's the time line in terms of when you expect to exceed the current 1 gigawatt of capacity?

So look, I think the exact time lines are not something that we would be discussing at this point in time. But here is the point, Michael. We pride ourselves on moving at AI speed. We are going to make sure that we are going to have capacity all of this year and all of next year to meet the kind of time lines that others can't or definitely what our customers want.

And so the capacity we built and how quickly we get that on, we were built for this moment. It is not like, you can take any factory of any other legacy technology and expand capacity at this rate. It takes deliberate thought process to have said, I need the supply chain ready. I need the manufacturing ready.

I need the equipment ready. We can move on a dime. We can increase capacity in months. And we can -- in most cases, I would venture to guess, provide that kind of additional capacity faster than a data center can stand up their own data center.

So this is where we are today. This is where we'll continue to be through 2026, and that is why we are doing this expansion plan.

Okay. Great. Maybe if I could just throw one more in. Is there any estimate on how much this will cost? Or how should we expect you to fund the expansion?

So we are funded for what we -- we are well funded for what we need to do in terms of going to 2 gigawatts. Round ballpark numbers, think about $100 million is how you should be thinking about this. And it will come spread over quarters. And we have enough. We are well funded for it.

And our next question comes from the line of Manav Gupta with UBS.

Congratulations on a strong quarter. I wanted to focus a little bit on the improvement we are seeing in the operating margin. I mean, you are at already about 7.1%. And if you look at your guide for this year, it's close to like 8.5%, I think the midpoint of it.

So making excellent progress on the operating margin. So help us understand what's driving it. And then when we look at 2026 or '27, would the target be to get to double-digit operating margin levels also?

Operating margin, yes. Manav, that's a great question. Look, as you clearly understand and you are looking at the numbers and you've been predicting our models. What you're seeing is tremendous fiscal discipline, right?

So we have a finance team here that works so well with the rest of our organization in managing how we spend, whether it's in OpEx, whether it is in how we procure and whether -- and also, we have a commercial team that's extremely good at making sure that our customers get great value at the same time, we get value for what we bring to them in terms of time to power and other issues. Between that combination and our cost reduction continuing, you should absolutely expect our operating income to keep getting better as we go forward. Will it vary quarter-to-quarter within a year? Absolutely.

That depends on mix and volume. That will happen. But what has helped us this year is we are more level loaded by choice, and you're seeing that in our inventory hold when you look at our cash, okay? You're seeing that.

That is a deliberate choice that we have made. And overall, for the entire year, we will reap the benefits of it by level loading that factory.

Perfect, sir. We were on the AEP call yesterday, and they said some very nice things about you and your scalability of your product. I'm just trying to understand, you can scale up that order. I think you're doing a couple of data centers through AEP. And as you work with them, do you continue to see more AEP, BE collaboration and more deployment of your product through AEP?

As you know, the service agreement was for 1 gigawatt. And if you just look at the AEP numbers and look at where their demand and where that gap is, we are hopeful that working together that we can fulfill that gigawatt and think about future agreements like this very soon. And we are working together. It's not just a hope.

There is a pipeline, and that pipeline for AEP is again robust, and we work with them. And you will hear more, I'm sure, in the coming months.

Yes. I would only like to say one thing conclusively. Two quarters ago, I think I asked you a question that this DeepSeek is happening and some data center providers are pulling back and you were very firm, this is a blip, do not look at it. Things will correct very quickly. And you had great vision and foresight. Absolutely, it was a blip. Nobody is even talking about it now. So thank you.

[Operator Instructions] And it looks like our next question comes from the line of Chris Dendrinos with RBC Capital Markets.

Yes. I wanted to ask about the value proposition you all bring. And I guess maybe specifically on the combined heat power solution. I know in the past, you've kind of spoken about it and it seems really appealing from an efficiency standpoint.

So I'm curious, are you deploying that with Oracle here? And if not, I mean, what's kind of the interest level? And where are you at kind of in that solution process with other potential customers?

That's a very good question, Chris. And here is what we see. In our pipeline, the interest as soon as our customers know that we can offer the CHP solution commercially is very high. So most customers we are talking to today their approach is, give us this quick time to power ASAP.

And within a few months, can we come back to you and retrofit the combined heat and power. The beauty of our solution is we are able to offer that, right? It's like adding an app on your phone. We can add the CHP as an app.

And that flexibility, that beauty of our technology is very appealing to them. So we are seeing that not just from data center customers, but from our commercial and industrial customers whose factories need to be air conditioned or whose factories need steam. So we are seeing this across the board in every area. And you are correct to point out from a value proposition wise, right, it is the equivalent of not needing 20% of your power in a data center, right?

It is the equivalent of not paying for it when you have taken care of your cooling with our waste -- with the waste heat as opposed to putting more electricity. That's a big deal.

[Operator Instructions] And our next question comes from the line of Maheep Mandloi with Mizuho.

This is actually David Benjamin on for Maheep. So I understand there's a robust demand and it looks like based on prior guidance, 40% in the first half. It looks like you guys are on track towards the top end of the guidance. I was wondering just the first part is like what would it take to raise the [ indiscernible ] bottom end?

And then secondly, are there any concerns with pushouts from the fourth quarter due to potentially ITC being available next year?

Great question. So -- or a 2-part question. That was nice of you to sneak in the 2 questions together. So let me answer the second part first, David.

On ITC, I should have been clearer in my prepared remarks, but let me try to be very slow, deliberate and perfectly clear so everybody understands this. Our customers have no gap from a timing perspective on ITC. They used to enjoy ITC last year through this year. They'll continue enjoying all of this year, and it will continue from '26 all the way to '32, okay?

There is 0 gap. There seems to be some confusion among people reading this. Yes, you're absolutely right. The BBB reinstated ITC for fuel cells starting January of 2026, okay?

However, we have secured enough volume under safe harbor. So our customers in '25 don't have to wait for '26 to be to avail of those credits. And therefore, there should be no penalty for any customer or no advantage to any customer to push out buying or installing their systems in 2025. Is that clear?

Yes. Crystal.

Okay. Now in terms of the guidance and where we are, we have reiterated the guidance. Where we fall in that range, why do we give you a range? It has nothing to do with our ability to ship the systems.

It's got everything to do with if the customer is ready. Many of these things are greenfield. They need to finish their factory or their data center on time. They need to be able to connect.

Gas needs to be there and like permits need to be there. So it could easily move out a couple of weeks on either side or a month on either side. None of these projects are in jeopardy of coming in or not. But when we recognize revenue will depend on those things.

And until we have clarity on that, we have to give you a range.

And our next question comes from the line of Chris Senyek with Wolfe Research.

So congrats on the Oracle deal. I was just curious, since that order was expected to be delivered within 90 days, there was no change in the full year guidance. Should we assume this is already embedded? Or would you need to announce additional deals in order to achieve those targets?

We have told you from the beginning of the year, and we'll continue to reiterate that statement that a portion of our revenue for the year would be -- would come from deals that we book, build, ship and recognize revenue in the same year. It's a wonderful thing. I just remember right after our IPO, it used to be 18 to 24 months on average to book something and then get a deal done, right? The pace at which business is moving now allows us to do that -- to shrink that cycle very well.

And we would be concerned about that shrunk cycle if there was reasons to believe that it's not a secular trend, and it's just a seasonal move. We don't see this as a seasonal move. And the fact that we have a secular trend where the pace of business, the velocity of being able to convert a deal to a booking happens faster is all fantastic from where we sit.

And our next question comes from the line of Dushyant Ailani with Jefferies.

Just the one on -- I guess, in the past, you've talked about opportunities outside the U.S. And I think you've also mentioned Taiwan in the past. Could you kind of briefly mention how those conversations are progressing and what opportunities you're seeing outside the U.S.?

Look, today, when we look at our business in general, roughly 30% of that comes from international, 70% comes from domestic, okay? And we expect to continue that ratio at least through the next year or so because as all of you know and follow from what's going on, there's tremendous action here in the U.S. market, okay? And so we see that. But we truly believe in the diversity of the market.

Not only are we continuing to show strength in Korea and keep that business going. We are developing new markets. I think we have mentioned to you Taiwan, Germany, Italy and the U.K. are the obvious next places we are looking at. And in all those places, we are making progress, establishing into a new market and being able to get the policymakers, the regulators lined up with something new to them, very similar to it was new to people in California and like New England when we started in the early days.

That process is going, but I'm very happy with the progress we are making in those places.

And our next question comes from the line of Colin Rusch with Oppenheimer & Company.

As you look out at the landscape of opportunities, are there situations where you could end up being used as temporary power for a couple of years and then have those servers move on to other locations? And are you starting to see any sort of incremental demand growth for longer cycle industrial, potentially chemical plants and other things that are looking to ramp and have power shortages as well as the data center opportunity you guys talked about?

Colin, that's a very good question. And the answer is absolutely yes, okay? And this is the reason, if you remember, we completely changed our -- how we install our systems from a go and pour concrete and install it on a concrete to on a skid, okay? Think of this as grid to go, okay?

We can just take these units on a truck and drop it in a site and then move that skid from one location to another with ease. And all it takes is 3 connections, the gas connection, the electric connection and the communications, which is wireless. So it's as simple as that. And here is another interesting thing, right?

It is even better than if you think of temporary power coming from combustion engines because you can't take a few blades and put it in one location and a few other blades in another location if you want to take 100 megawatt and break it into 5 -- like 20 megawatts. With our Bloom systems, the modular nature of it, not only can you take our units and put them in different locations, you can fragment it or aggregate it as you wish. And so it is definitely a very attractive option that we can offer. And even the hyperscalers are very interested in that because in some locations, if they're able to get power, they can easily move it to another location where they need more power, right?

So this becomes very, very attractive. In fact, to us, this is an amazing selling point. And are we concerned about will they need the power 2 years from now? Anybody who thinks 2 year is the bridge and something miraculous is going to be there, you're just going to go to the next bridge.

It's a bridge to a bridge. It's not a bridge to a solution, and you all know that.

And our next question comes from the line of Sherif Elmaghrabi with BTIG.

Another tax credit question. Safe harbor aside, the BBB does give more visibility. And my understanding is that domestic content bonuses have actually been raised. So my question is, with the tax credit picture is set, hopefully set, does that put you in a position to push pricing?

Sorry, something got cut off on this. I'm like really sorry on our end. Can you just repeat that question one more time, if you don't mind, Sherif? Sorry.

Sure. I'm saying that BBB crystallizes the tax credit picture. And I think in some cases, it actually increases the benefit. So does that give you -- does that level of visibility put you in a position to push pricing?

Yes. So what happens with the BBV is there is no domestic content adder when it comes to the ITC, right? So I think I'm answering your question. If not, let me know, and I'll answer it.

So it is a flat 30%, okay? Whereas in the previous version of the bill that ended last year, but safe harbored now, our customers can avail of either 40% or 50%, depending on whether they are not in an energy community or in an energy community. If they're not in an energy community, it's 40%. If they're in an energy community, it's 50%.

So from that perspective, yes, their subsidies go down a little bit. But given how high the price of electricity has gone up, at 30%, our attractiveness will be extremely high. And you also know that every year, we bring down cost reductions between those 2 things, we don't see any issues with relation to maintaining our margins, if that's your question.

And our next question comes from the line of Noel Parks with Tuohy Brothers.

I wonder if you could talk a bit about product development and maybe update us on your upcoming generation of the Bloom Energy server and what incremental benefits you're anticipating from that? And any comment you have on the effect on the economics for the product for you would be?

No. Thank you very much for that question. So here is what is really interesting about the Bloom platform going forward, okay? That improvement that we are talking about does not need to happen by generation as a step function.

It is happening continuously quarter after quarter of things coming of new ideas being robustized, completely manufacturable and entering into production in a line. So think of our product continuously improving as opposed to going from one model of a product to another model of a product. And we are able to incorporate that seamlessly. So lots of developments are continuing to happen.

And in a way, if you saw our earnings script, we are maintaining guidance on our margin even though there can be a 4% tariff fit on our materials, right? All that is coming from further improvements coming into our product. So we don't anymore talk about our product in terms of generations, but in terms of it is getting healthier and healthier as the days go by, and it's getting better and better. Why is that?

Again, it goes back to -- if you look at my script, we get real-time feedback from the digital twins that we have set in place, the 4 trillion data points that come into us and how we analyze it, how we learn from it, how we improve it, gives us a unique ability that I don't think exists in the power industry other than at Bloom where we are able to improve our product. So this is -- you will see this as a continuous improvement as opposed to a step improvement going forward. And we bring new attributes. So for example, CHP was a new attribute.

Load following was a new attribute. Being able to operate island and load following was a new attribute. So we bring these attributes into the product.

And our next question comes from -- it looks like we've got another question from Chris Dendrinos with RBC Capital Markets.

I wanted to go back to the value proposition, and I think you slightly hit on this in Colin's question. But maybe just comparing your solution to like a natural gas turbine, we've seen those get deployed in some situations. And I think -- could you just help us, I guess, paint a better picture of where you see your solution really being the most optimal compared to a gas turbine. I mean, to me, it looks like it should always be, but there's certainly reasons that are driving some customers to take up a gas turbine.

And so just trying to get a sense of what those kind of driving forces might be.

Chris, I completely agree with you, but nobody gets fired for buying IBM, right? So anyway, if you look at economics, here's what I can tell you, you look at the demand growth and you look at where numbers are going, what we heard even 6 months ago of 20% of data centers may be islanded and not connected to the grid, that number is now hovering on 40%. The amount of data centers that are telling us in the surveys that they would use islanded power is going up. When you go there, the monolithic turbine needs additional monolith to be able to -- so when one unit is being serviced, the other unit is on standby.

You put that together compared to the Bloom architecture where it's LEGO block and you compare apples and apples for a grid level availability of 999 or even a high 9s available of 99999s or 999999s you compare that apples and apples, our CapEx compares favorably or at a minimum at parity with turbines. On the other hand, those turbines have at least 15 to 20 percentages -- like percentage points more fuel that they will consume compared to us. So on the OpEx, there's a significant win. And on top of that, we have no air pollution, whereas getting an air permit to put a lot of turbines if you live in a populated area is very difficult.

You're probably reading in the press. So you combine all those things, A, because we are easier to permit, we remove the friction to permitting, so we are faster. Time to power is everything in this business. Secondly, operating cost is lower.

CapEx is at parity. You put them all together, I think we compare more than favorably to any other alternative way of producing electricity.

And our next question comes from the line of Skye Landon with Rothschild & Co. Redburn.

More of a clarification from myself. On the AEP announcement recently regarding the Amazon Web Services and the other company, are you able to confirm whether those projects are included within the 100 megawatts or if those are additional on top of the 100 megawatts?

It is part of the 100 megawatts that was in the PO and the other 900 megawatts are what we are working actively on the pipeline.

And our final question today comes from the line of Dimple Gosai with Bank of America.

I actually have 2, if that's okay. The first one is just trying to get more clarity or understanding around the Oracle deal. Is this a done deal? Is there a framework to deploy? Any KPIs that need to be met? Any color would be super helpful for us. And the second question is if you can talk a little bit about your capital needs or liquidity in terms of funding the manufacturing expansion.

So on the -- first question on the Oracle deal, we are a company that don't discuss on earnings calls, LOI and MOUs and things like that, okay? This is a purchase order we are executing. They will have power in 90 days period. In terms of how much capacity and dollars we need, you may have joined us late.

I don't want to hold everybody up on that. We have already answered that question. It will be in the transcript.

Thank you very much. And with that last question, let me say that, look, AI is moving faster than any technology in history, faster than the Internet. And of course, orders of magnitude faster than the electricity industry. And it's demanding more, more power, more data centers, more urgency.

Our company Bloom, our entire company was built for such a moment. We are ready, and we can move at AI speed. The advantages of our purpose-built digital platform has never been more obvious and relevant to everybody. And the opportunity in front of us is both massive and secular.

We are confident in our strategy. We are confident in our execution. And internally, we ourselves have shifted to a higher gear because we are going to blaze forward. We thank you for your confidence in us and your continued support.

We wish you a good day.