3 Semiconductor Stocks To Buy for the Second Half of 2026!!

Hey everyone and welcome back to the channel. I'm Rachel and I'm joined today by Jose. And today we're going to be talking about three stocks that we think are great buys for the second half of 2026. You know, the first half of this year, it's given investors a wild ride. We've seen a lot of volatility across a wide range of industries, notably the semiconductor space. We're seeing astronomical data center spending guidance from the hyperscalers. We're getting a lot of mixed signals coming out of the Federal Reserve regarding interest rates. So, there's a lot for investors uh to be looking at right now and understanding as it pertains to their long-term portfolios. But, as long-term investors, we know that looking in the rearview mirror doesn't help you build a portfolio that stands the test of time. So, we have to look at what's coming next. Uh, and right now we think the second half of this year is shaping up to have some of the biggest enterprise hardware rollouts and infrastructure spending shifts that we have seen since the AI boom kicked off. So today we're going to be talking about three top stocks that we think look like compelling buys right now. Let's get started. >> Now before we continue with today's episode, if you want market beating stock picks from our analyst, make sure to check out the pin comment and the description. Using that link gets you a promotional offer as our thanks for being a viewer. Thank you and let's get back to today's episode. >> All right, Jose, let's jump right in. The first stock on our list today, this is a semiconductor giant that's been playing second fiddle in the headlines, but they are about to take a big swing at the reigning king of AI hardware. Of course, we're talking about advanced micro devices, ticker AMD. >> Yeah. Yeah, Rachel. I mean AMD is has had an amazing first half of the year in regards to stock price but I think the market is still underestimating how strong the second half is going to be and and the main reason is the second half of this year for advanced micro devices is going to be a tipping point for them in a new product which they call the Helios rack. Now before we understand the Helios rack I think it's important to understand a little bit of how the market is shifting. So prior prior uh the AI data center just the data center market in general you used to just buy a chip and then the company that bought the chip would kind of either hire someone or work with somebody to build a server around that chip and that's what was needed. But now as AI workloads and AI demands and other clouding cloud computing workloads increase in demand, you also need better infrastructure and infrastructure that has become more of an ecosystem. So AMD and other players like Nvidia are now designing full rack scale solutions where they pretty much have a design process for the whole server. AMD's Helopra, which is going to come out in the second half of this year, is going to be the first time they actually do this in the AI market. This is going to include their AI GPUs. Uh the MI450, it's going to include their AI CPUs or just their overall CPUs for for regular workloads. It's going to include networking solutions. It's going to include DPUs and so much more. So AMD is finally entering into this rack scale with the Helios rack. And then what's even more important about this is they already have a nice amount of customers backing this Helios rack system. Uh these are big players like OpenAI. They recently announced Anthropic. Microsoft just mentioned an extension that they're officially signing and deploying the Halos Rex system into their data centers. You also have Meta who has made a huge announcement with AMD and Oracle also has uh in the works the MI450. Now the thing is many investors might say Jose that's only five customers. What what that that's not a lot but these five customers each have focused on over 1 gawatt of data center in theory. In theory to build the gigawatt of data center is somewhere around $50 billion rough or take 10 mill billion depending on the and it's not just all AI chips it's the chips it's the data center itself the real estate the power solution and all that but if you have one customer building data center solutions out of a gigawatt scale you can see the massive amount of revenue opportunity coming for AMD and that is expected in the second half the first servers are expected to ramp later in September. Uh, and you expect to see that ramp increase throughout the upcoming quarters as well. And that's why Rachel AMD is one that I'm really excited about for the second half of this year. >> Well, I think one of the things that's really interesting when we think about this, I mean, the dominant narrative on Wall Street for a while has been that Nvidia has an unassalable monopoly here, right? Because they own the software ecosystem. They've obviously got the hardware lock in. And you might look at the revenue gap between Nvidia and the likes of AMD and assume that AMD can't compete. But what you've just highlighted with the Helios system, I think it really does disrupt that thesis. I mean, they are delivering a preconfigured rack scale system that really matches that full stack hardware integration and that removes a lot of the integration hurdles for the big cloud providers, right, who are desperate for a secondary source of supply. I mean, you you named a lot of these companies, the the Microsofts, the Oracles, the BetAss, you know, the list goes on. When these companies are looking at their capex, they don't want to be single sourced to just one chip provider. They want pricing leverage. They want volume. And I think the thing that really sticks out to me is that AMD's Helios gives them a a plug-andplay alternative, if you will, that specializes in inference. And that's really the direction that the software industry is moving as companies try to monetize uh their AI features. So I do think that creates tremendous long-term growth opportunities for AMD that you know could be profitable for the business, profitable for investors as well. >> Yeah. Yeah. Rachel, and just before we move on to the second stock, I I I talked a lot about the bullish points for AMD, right? Kind of this new Rexdale solution. But one thing that I do want to warn investors is there is risk with this, right? Building a scale system is extremely complex and there could be some issues that might arise in the short term of things because maybe there might be an issue of something they overlook. I would say the market is going to overreact to news like that. The engineering complexity of a rack steel system is not something that you can just do right off the bat. It's not impossible, but there is the opportunities for some hiccups. And and so so now that I talked about the risk there, I I want to jump into the second stock. And obviously it fools, if you have me here, if you have me here, I'm going to talk about another semiconductor company. So >> I'm shocked. How can this be? >> But this one, Rachel, it's not one I talk about too often. And I think fools are going to be like, Jose, another AMV or let me guess, Nex is Nvidia. No, it's not Nvidia. It's actually Qualcomm, ticker Q. So when most investors think about this uh they think about Qualcomm in the mobile business, the phone company, they make the Snapdragon chip that goes in a lot of Android devices that go into a lot of non-Apple phones. It's kind of like the top CPU in the mobile market outside of Apple's own internal uh CPU, but I'm not really too focused on the mobile business for Qualcom. Uh, obviously with memory pricing, we are seeing kind of a weakness in the consumer space in in certain electronic goods. I'm not going to lie, Rich was just looking at phones because mine broke and I'm a little I'm a little taken back by the prices that some of these phones have gotten. But my thesis here for Qualcomm is the second half of this year. They're entering into a market that they virtually had no exposure to in the past, and that is the data center play. designing their own AI CPUs for the data center market. They're also designing their own AI chips for the data center market. And while all of those are expected to happen in further years, there is a market that they're going to be opening up in the data center space this year uh more importantly in the second half and that is the ASIC plan. They are helping two hyperscalers design some form of AI chip or custom solution for their needs. and we are expected to get revenue out of that in the second half of this year. Now the reason I put this in this bucket is because many people don't look at Qualcomm in the data center play. I mean you have so many names. You have AMD, you have Nvidia, you have others that we might be talking about in the future, Broadcom, Marvel val um the list can go on. So rebris. So when a new entry and a new player comes into the space, the market goes but can you succeed? Can you actually make this a market for you? Would you be able to collect revenue or are you just giving us a press release and expect us to buy up the stock? I personally believe because of the shortages in this whole AI industry because there's not enough compute, players don't mind going with player number three, player number four, even player number five because they just need more and more compute for these AI workloads. So for that reason I am adding Qualcomm here on the second half of this year not necessarily for the mobile space but more for that data center play that they are finally entering into. >> Yeah I mean there there's also a few thoughts I have here but I think they've shown that they're very serious about this data center push right you know they completed a nearly $4 billion acquisition of by a software infrastructure company called Modular Inc. Right. And and these Dragonfly C1000 CPUs, they target agentic AI workloads, which is of course what we are hearing about as sort of the next generation of the the AI revolution. So basically, you know, instead of buying a generic power hungry GPU that is built to do everything, the hyperscalers want custom silicon that can handle, you know, sequential reasoning, context switching. What's interesting is if you look at Qualcomm's business model over the last few decades, you might be thinking, well, what what have they been known for? They have been known for designing these very complex processors that can handle huge data workloads while running on a tiny smartphone battery without overheating. And it's interesting to see how they're taking that expertise and transferring it over to the current needs of the AI industry. Because of course right now the entire AI data center space is facing a huge electricity and cooling crisis, right? I mean power consumption is the number one operational expense. It's probably the most significant bottleneck that we're seeing in the AI buildout. And we're seeing from these numbers, these early numbers that Qualcomm's mobile architecture delivers about two times better performance per watt than existing server CPU benchmarks. I think that's also something that investors, you know, need to pay attention to. And on their recent investor day, uh, they raised their 2029 non-handset revenue target to $40 billion. That's up from their prior projection of $22 billion. And of course, you mentioned that Meta partnership. So it is really interesting to see how kind of a slower growing business known for its expertise in the smartphone space is now leveraging that experience and background and those resources into the the next wave of the AI revolution. >> Yeah. And again I just want to go back to the risk. One of the biggest risk is this is a new play for them, right? They they've done really well. They did some success in the mobile business, the data center market. They were giving us those numbers. guidance increased dramatically for the fiscal uh for fiscal year 2029, but we have yet to see any data central revenue just yet. And that's what the second half is going to tell us. And if it does give us those great answers, Rachel, I think this is going to be a nice pick for for the semiconductor investors. >> Yeah. Well, before we let you guys go today, I'm going to hit one more stock, our third stock today. Uh I want to take the lead on on a company that I think really bridges the gap between Jose's first two picks, AMD and Qualcomm. You know, if we're sitting here, we're bullish on uh the roll out of AMD's Helios rack systems, then we're also bullish on the explosion of custom enterprise silicone for hyperscalers, then I think one stock that also looks like a really solid buy right now is Broadcom, ticker AVGO. Broadcom is what I would call very much a pick and shovel way to invest in the AI hardware ecosystem. And one of the things that's interesting is how they actually connect directly to AMD's new Helios system, which Jose was talking about earlier in the video. You know, you can have the most expensive GPUs in the world sitting inside a server rack, but if those chips can't communicate with each other, instantly with zero latency, the system bottlenecks and it essentially becomes useless. So, Broadcom has worked really closely with AMD to design the the scaleup switches, the UA links, uh, that physically bind the Helios system together. So that you know proprietary fabric is really the invisible technology that allows thousands of separate chips to communicate so that they act as this cohesive supercomputer. So as the helio systems begin shipping to customers like Microsoft Meta and others that we talked about getting into the second half of the year and into next year Broadcom is also going to be a direct beneficiary of every single rack deployment because compute is essentially useless of course without the networking bandwidth. But the other thing is the the ASIC trend that Jose mentioned earlier as well. I mean Broadcom is the global market leader in co-designing custom AI chips. They already hold dominant relationships uh with Alphabet right for their TPUs and and Meta for their custom silicon. We're also seeing in latter half of this year a third major cloud hyperscaler is ramping up high volume production on brand new custom ASET codeesign by Broadcom. So, you know, we're seeing signals for management that the AI revenue is going to be continuing to tick up, mostly driven by this custom silicone ramp. I think that this makes them a business that is a really intriguing play on these AI trends, but also importantly, I think this is a business that is really wellrun, and I think it's balance sheet backs that up. >> Yeah. Obviously, I'm I'm happy because it's a third semiconductor stock. So, it's it's like a a dream episode for me, but I like the angle play that you had there, Rachel, because normally when we think of Broadcom, the first thing everybody talks about was the ASIC business, and I know you brought that up, but the networking side is no matter who wins in forms of chips. It it doesn't even have to be their TPUs. But if all these open- source solutions start to become favorable for hyperscalers or AI chip designers, then Barom is going to have that massive opportunity in networking. I really like this play, Rachel, and I hope the fools are enjoying these three companies, right? Well, we talked about three great businesses today. Uh AMD, Qualcomm, and Broadcom. They all touch different corners of the tech sector, the AI infrastructure market, but they're all very much tied to long-term tailwinds that are not only going to be playing out over the next 6 months, uh, but into the end of the decade. We do think it is going to be a really exciting second half of the year to watch these businesses play out and we'll be listening into the quarterly earnings calls to hear more of what management has to say. Uh, but we want to hear from you guys now. You know, let us know in the comments below which or any of these three stocks you own in your portfolio if you want to own them or if there's another stock for the second half of this year that's sitting on your watch list right now. Hit that subscribe button and give us a thumbs up if you enjoyed the breakdown and we'll see you in the next video.