Oracle's AI SURGE: Is NVIDIA's Stock About to Skyrocket?

nvidia's stock is down over 20% from its year- to-day high as many worry that the AI Market is filled with hopes and dreams however Oracle just reported earnings giving us a great reminder of how strong and resilient the AI Market truly is especially noteworthy is the CEO's mention of Designing a massive data center for another company one that already has building permits for three nuclear reactors how many Nvidia gpus do you think three nuclear reactors can power if you think it's more than one make sure to hit the like And subscribe button now in today's episode we're going to cover the following first a nice top level overview of Oracle earnings second a discussion about AI training versus AI inference and the potential of overbuilt issues the Intriguing nuclear power story data center builds from Oracle commentary on AI demand and my final thoughts on these developments so let's get started I want to thank the mly fool for sponsoring this video and check out fool.com hose for the 10 best stocks to buy now with that link you get a promotional offer for the subscription service now let's continue with today's episode now Oracle is up roughly 8.88% as I am recording after hours as the company did give some amazing earnings and for those not familiar Oracle right now is trying to improve and increase their clouding infrastructure solution which is obviously doing really well with the AI market now a quick top level overview on earnings earnings per share were $1.39 versus $1.32 expected Revenue $13.31 billion versus 13.23 billion expected so a nice beat in earnings per share and a modest beat in Revenue now for the current quarter Oracle expects Revenue to grow somewhere around 8 to 10% pretty much in line with what analysts were expecting and in forms of earnings per share again pretty much in line with what analysts are expecting now the main reason the company is up after hours is because the company continues to see strong demand for its AI Solutions so next I want to talk about training versus inference and one of the big things to understand is when you're looking at training training needs huge GPU clusters now when you do inferencing you don't need the same type of computational power as training and this has created kind of a bearish thesis that eventually we're going to be done training models and we're only going to be doing inferencing and if that's the case there's the potential for companies like Nvidia and all these cloudy solutions to either lose out on Revenue because they no longer need that computational power for training um so they it's more like hey are you overspending too much in training infrastructure when the market will eventually move into inference and the CTO of Oracle actually gave a very very amazing answer in my opinion he first kind of talked about the human life in human right you would think training would be like going to college you go to four years to college that's you done training once you're done training you go out to the real world and get a job and in theory that should be influencing but just like in real world even after you're done with college or your commun or whatever type of education you keep learning more and more throughout your life so training is going to be the same way training is going to continue continue continue to go on the next thing is right now Frontier models which are kind of like these gpts this llama 2 llama 3 llama four llama 5 they are going to continue to train the Next Generation right right now we are training I believe llama 4 GPT 5 then there's GPT 6 and he mentions is that this is not going to stop anytime soon and it's pretty impressive because to be able to train these Frontier models it's about a hundred billion do price Target let me repeat that to train these AI models is about a100 billion price Target to be competitive as one of the top players so that's in so right off the bat we know that there's only going to be a limited amount of companies that are going to be able to really hit in this training market and I think that's why we're seeing Google we're seeing Microsoft and we're seeing meta invest a lots and a lots of money on AI Solutions now he mentions that hey look while he also mentioned to kind of ease these worries that the next Frontier models are going to continue to get upgraded and there's going to be a next new frontier your model there's also going to be a lot of specialized models for example training uh uh kind of getting a model or or training a solution to be able to look at biopsies of slides or CAT scans to Discovery cancer that's something that doesn't get trained in llama that it's not something that gets trained on GPT 5 that's a very unique model that is specialized for a very unique solution and very unique data sets there's also other samples like that in the healthcare Market where you can use blood test to discover certain types of illness so the CTO of Oracle believes that this is going to continue to be the case where while obviously we're going to see a few a handful of players do Frontier models we're going to see many many companies create unique Solutions like this right I just mentioned two unique solutions for the healthcare Market but there can be hundreds in the healthcare Market alone then in things like Automotive like robotics like software like chemistry the list goes on and on so there's going to be a lot of specialized models in the future now he kind of mentions that if you're talking about training and inferencing if you believe that we're going to move from training to inferencing anytime soon you're many years ahead he believes that even in the next 10 years we're going to continue to battle this AI race and for that we're going to continue to invest in training Solutions so he mentions that the AI training space is pretty much just growing larger and larger and there is no slowdown or shift coming so obviously it's kind of important to remember that the CTO obviously benefits and this company benefits from training so they're going to give as much bullish case as they can but I do believe there is some truth here and this again boats well for a company like Nvidia where it shouldn't be seen as a bearish case that training is going to slow down anytime soon in theory the computational power needed for the next Frontier Model is just getting higher and higher so in theory more AI Sol more AI chips are going to be needed now the second topic I want to talk about is that crazy AI story with nuclear plans and this was kind of a follow on to this training versus inference um the CTO just wanted to kind of paint a picture of how this Market is so right now he mentions that they are in the middle of Designing a data center that's north of a gwatt of power a gwatt of power is over 1,000 megawatt and the current biggest one that I we have discussed in public I believe from Oracle is something around 800 megawatts so we can see this is much much bigger and he mentions that they've already found the location in the power place and when they looked at it the other company already has building permits for three nuclear reactors these are small nuclear reactors but it kind of just showcases how crazy the market is and also how power limited we are so if we're thinking about three nuclear reactors this is going to be a lot of AI chips that are going to be powered the other thing to remember is these are data sensors that are being built for the future it's going to take probably at least 5 years to kind of have these Solutions up and running so it's this is not going to be for nvidia's h100 it's not going to be for nvidia's h200 it's probably not even going to be for nvidia's Blackwell or Ruben platform this is probably going to be for the platform following Reuben so it just kind of showcases how if you are an Nvidia investor we are still in early Innings based on what the CT the CTO of Oracle just mentioned now I also want to talk a little bit more about Oracle building their data centers they mentioned that right now Oracle has about 162 cloud data centers a mixture of some are live and some under construction I believe if I'm remembering my numbers correctly is about 85 are live and the other 77 are under construction so we can see this company is building more data sensors to keep up with the AI demand one of the largest data sensors that they are building is a 800 megawatt beast and will contain let me say this Acres of Acres of Nvidia GPU clusters to be able to train the world's largest AI models so Oracle just showcases that one of the main reasons that they're winning a lot in AI training business is because they're doing really well at being able to build 32,000 node Nvidia GPU clusters something that that um it's very very hard work but they're able to do it really really well and the main reason they are winning in their training business so again all I continue to see is a lot of bullish cases for NVIDIA now some final things on earnings before I share my overall final thoughts they mentioned that capex is going to be roughly double in fiscal year of 2025 versus 2024 as they continue to invest in in expanding their AI capacity or their Oracle clouding infrastructure in quarter 1 which is normally a slow quarter for them it was actually a positive quarter for them where they signed 42 additional Cloud GPU contracts for a total of $3 billion so overall again Oracle continues continues to give us bullish scenarios for NVIDIA so even if you're not an oracle investor I think this did give us a lot of great information of the future for NVIDIA and for the AI Market it really kind of shows us how we're still in early Innings now my thoughts on Oracle Oracle luckily for me I did enter in the very very very very low 100s and I personally am happy with the stock being up as much as it is I have no intentions of selling as I believe the market cap is going to be much larger in the next few years so I would definitely like to hear from you do you believe you would buy Oracle at these levels would you wait for a buying opportunity or do you think this company can do much better in the future take care have a good day and see you next time