Broadcom faces short-term technical weakness and soft guidance, but offers a potential entry point at support for long-term AI growth.

First one here is going to be Broadcom and the ticker on this one is AVGO and it was 2.41% lower in the pre-market on Thursday as it reported better than expected financial results for the third quarter of fiscal 2026 but issued weak guidance for the fourth quarter.

let's start talking about some of these earnings and we're going to start with Broadcom. I know this is the one that everybody uh is probably watching the most. First of all, if we take a look at the chart, kind of looks like a precarious position to be quite honest with you.

I'm definitely bullish on AI, especially for the long term. And the Broadcom report kind of lends itself to long-term bullishness on AI. But in the meantime, the chart is precarious because we traded up off of a low.

We actually broke a sideways area. We had a break of structure to the downside, but it was very very temporary. We traded back up to the 200 day. We actually consolidated along the 200 day, which can be a great thing if the next move is up, but the next move wasn't up.

We are gapping down by 3.78% here. And as you can see, we're actually rechallenging this low. And if this breaks, you can make a case that Broadcom's got a number of different things that it's going to have to take care of before it moves back higher.

Namely, we have this small gap in here that fills down at about 334. We also have a fib level in here about 331. And we have a volume point of control. Actually, let me make that a little bit bigger here.

Uh we want to scroll back and encompass all of 226. We have that volume point of control just slightly above that fib level. So 331, 336. That area in there might catch us. Um that's the gap fill, fib level, and volume point of control.

And then furthermore, we have the round 300 number coming plus the 500 day at 290 and a half.

So if Broadcom continues to sell off and you have a long-term bullish view here, you might be able you might be able to pick this stock up at a discount on a support level. So keep that in mind if you have a long-term perspective on any of this stuff.

The Broadcom CEO says the AI demand could be significantly higher, but land, power, chips, and substrates are holding back data center deployments. You know what's funny about this is that we heard this from Dell.

They had an monster earnings report traded higher and told you that they were constrained by power and chips.

Broadcom reported fiscal third quarter revenue of 29.59 billion that beat the consensus of 29.36 billion while adjusted earnings of 332 per share also beat the estimates of 324 per share according to Benzinga Pro.

However, the company's fiscal fourth quarter guidance seems slightly underwhelming to investors following which the shares fell nearly 4% in the pre-market on Thursday. The action actually started in the after hours yesterday.

Broadcom is positive on generating billions of dollars in AI revenue, but believes issues around supply chain constraints and the actual deployment of data centers remain. That might be one of the headwinds here.

CEO Hawk Tan said the company's fiscal thirdarter earnings call that quote we're very careful and to be honest we try to be conservative so we're giving you an outlook and yeah demand we can ship significantly more end quote.

So I'll tell you what that quote there just kind of kind of makes me think like is it seems like there's something coming down the pipe. It seems like he's telling you they're trying to be conservative. the demand is really good and things might be coming down the pipe. So, just keep that under your head.

On Wednesday, Broadcom said that it expects AI revenue to double to 115 billion in fiscal 2027 and a double again to 230 billion in fiscal 2028. This is supported by secured supply lines and customer demand.

Tan went on to say that quote certainly our customers want us to ship more but we're have secured but we have secured supply and we think it's the right number to put it to 115 and if circumstances change in our ability to scale more supply change of course we will uplift end quote.

Ken also addressed other concerns related to LAN power and shell or LPS when it comes to data center deployments and potential impacts on the outlook. quote, "It is a multi-dimensional issue to get an AI data center deployed."

Yeah, we now are more sensitive about, especially at scale, and we are now delivering at scale to our customers. XPUs, AI data centers built upon XPU computing accelerators, land, power, and shell.

As I indicated, the question with this will the question with this is will there be will it be a big concern? It's more than a big concern. it dictates specific timing of when this capacity gets deployed and will be available. End quote.

Tan also talked about high bandwidth memory or HBM chip constraints which the company does not supply but its customers secure for their needs as a bottleneck. He said quote and of course we all know about memory HBM memory and beyond HBM memory the system memory that goes into AI servers which we don't supply necessarily but all our customers have to secure too.

So this is a multi-dimensional problem coming from various sources and then depending on different times each might become a bottleneck end quote.

So, I find that interesting because he's basically saying, "Yeah, we're doing everything we can, but our customers who represent an absolute ton of our revenue also need another product that is supply constrained."

So, it's not even their fault. That's what I'm gathering from him. I guess I'm putting my own opinion on it, but it seems like he's saying it's not even their fault that they can't meet these, but again, thinks that they will be able to in the future.

Gary Black, the managing partner of the Future Fund LLC, in a post on X said, quote, "Broadcom down 7% after hours after a Q3 results beat, uh, but Q4 revenue guidance fell slightly short of analyst expectations."

That's 34.8 billion versus the 35 billion that was expected. He goes on to say, "As the AI rev strength becomes clearer to investors, I expect the stock to recover somewhat."

Jim Kramer sees Broadcom's business exploding as CEO Hawkan says custom AI chips beat Nvidia's Grace Blackwell at half the cost.

Uh-oh, trouble for Nvidia. We'll see. Despite weak fourth quarter guidance, Broadcom, ticker AVGO, their management revealed a multi-year AI revenue outlook and touted its custom silicon, which Mad Money host Jim Kramer dubbed as an upcoming quote explosion of business end quote.

Meanwhile, CEO Haktan claimed in the earnings call that Broadcom custom chip for Open AI platforms or excuse me for Open AI outperforms Nvidia's flagship Grace Blackwell for significantly less money.

Despite a double beat in the third quarter, Broadcom issued soft guidance expecting fourth quarter revenue of approximately 34.8 billion versus estimates of 35.03 billion.

Looking ahead, TAN mapped out an aggressive growth trajectory, projecting AI semiconductor revenue to double to approximately 115 billion in fiscal 2027 and then double again to 230 billion in 2028.

Man, Enthropic is slated to deploy 5 gawatt of compute in 2027 and another 10 gawatt in 2028, cementing its status as Broadcom's largest custom chip customer. While Open AI plans to deploy 1.3 gigawatts of Broadcom's custom jalapeno chip next year. Jalapeno chips.

Following the management's comments on Broadcom's trajectory, Kramer added in an expost that quote Broadcom tells a story of an explosion of business that is coming. End quote.

So again, I hark back to the inverse Kramer phenomena that some of us subscribe to. Is Kramer right? Do we see Broadcom having a quote explosion of business? That would be good for the stock.

I have to believe. Or did he just mark the top? Did he or I I guess I should say did he confirm the top?

On the earnings call, Tan took direct aim at the current GPU market leader. He highlighted Jalapeno, OpenAI's first generation custom accelerator built by Broadcom. Tan stated that Jalapeno outperforms the Grace Blackwell Ultra in performance per watt, latency, throughput, and power for inference workloads.

Pointing to the economic advantage of specialized hardware, Tan noted that a chip optimized for specific large language models will quote outperform any GPU end quote. And can do so at half the cost of a GPU, also end quote.

Broadcom shares initially slumped over 5% after hours after the week fourth quarter revenue guide. However, the stock has recovered as the market digested a historic surge in custom AI chip demand.

Third quarter AI semiconductor revenue hit 16.7 billion. That was up 221% year-over-year. Total third quarter revenue reached a record 29.6 6 billion with non-GAAP earnings of 332 per share beating expectations and the dramatic turnaround in sentiment prompted Kramer to call the print an overtime save.

Kevin here says, I calculate its 12.5 forward PE. Do the math.

He says, "Lion says, I said short Aago since creamer talked about laughter really is the best medicine."

Broadcom has continued to drift a little bit lower. We actually have lower lows in the pre-market. So, we want to see if this trend stays intact. Ultimately, your probably se 342 but maybe even 346 might be a support area if Broadcom continues to fall. this down 3.7% here today.

And that's great because if you want to take a look at Broadcom, we can. I already covered that with the news and the chart earlier, but we can take another look at it.

Jay here saying Broadcom that report reminds me of Marll. It was okay but not great. So Jay, what was interesting is that they said something very similar, right? They said that they were constricted by power and and and memory which was also which was also similar to Dell and Marll.

Now interestingly enough the Broadcom outlook was really good. So the question that I posed before you uh joined here while you were down the dial the question I posed to the group was is Broadcom is this move down especially if it comes into a support area is this a buyable move down because if you believe the long-term view and what they're saying about their guide this should go higher no

regarding my Broadcom question of do you think it's good long-term? Jay says no. He thinks it's dead money.

Ryan, it's a great opportunity to buy Marll Arm and Broadcom by dollar cost averaging and hold until the end of the year. I think you'll be rewarded big.

Broadcom is still relatively weak. Thank you Trey for the reminder there. and Broadcom still holding its lows for the moment.