# Beta Finch Podcast Script: Texas Instruments Q4 2025 Earnings

**ALEX:** Welcome to Beta Finch, your AI-powered earnings breakdown. I'm Alex, and I'm here with my co-host Jordan to dive into Texas Instruments' fourth quarter 2025 results. This podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

**JORDAN:** Thanks Alex. Texas Instruments just delivered some really interesting numbers that caught a lot of people's attention. The big story here isn't just what happened in Q4, but what they're projecting for the first quarter of 2026.

**ALEX:** Absolutely. So let's start with the headline numbers. TI reported $4.4 billion in revenue for Q4, which was up 10% year-over-year but down 7% sequentially. That sequential decline is pretty typical for the fourth quarter. But Jordan, what really stood out to you?

**JORDAN:** The guidance is what's really fascinating here. They're projecting Q1 revenue between $4.32 billion and $4.68 billion, which would represent the first sequential growth in a first quarter in about 15 years. That's a huge departure from normal seasonality where you'd typically see a decline.

**ALEX:** That's remarkable. And when you look at the segment performance, you can see why management is optimistic. Their industrial business was up 18% year-over-year, automotive grew in the upper single digits, but the real star was data center - up around 70% year-over-year.

**JORDAN:** The data center story is particularly compelling. CEO Haviv Ilan mentioned they're now at about $450 million per quarter in data center revenue, and this market has been growing for seven consecutive quarters. They've repositioned data center as one of their five key end markets, which tells you how strategic this has become.

**ALEX:** Speaking of strategic positioning, let's talk about their manufacturing investments. They're nearing the end of what they called a "six-year elevated CapEx cycle." Rafael Lizardi, their CFO, mentioned they expect CapEx between $2-3 billion in 2026, but here's the kicker - with the new 35% investment tax credit from the CHIPS Act, they're getting significant offsets.

**JORDAN:** That's huge for their economics. They're also expecting up to $1.6 billion in direct CHIPS Act funding as they hit various milestones. But what I found most impressive was their free cash flow story - it nearly doubled to $2.9 billion in 2025, representing 17% of revenue.

**ALEX:** And they're returning that cash to shareholders aggressively. They returned $6.5 billion over the past twelve months through dividends and buybacks, plus they increased their dividend by 4% - marking 22 consecutive years of dividend increases.

**JORDAN:** Now let's dig into what's driving this unusual Q1 strength. Management was very clear this isn't about pricing - in fact, they expect overall pricing to be down low single digits, which is pretty typical for them. Instead, they're seeing genuine order strength.

**ALEX:** Right, and during the Q&A, executives mentioned they saw orders improving throughout Q4, with stronger month-to-month progression and building backlog. They're also seeing elevated "turns business" - customers wanting immediate shipments - which suggests real underlying demand rather than just inventory building.

**JORDAN:** The industrial recovery story is interesting too. Even with that strong 18% growth, Haviv Ilan pointed out they're still about 25% below their 2022 peaks in industrial. So there's potentially a lot more room to run as that market normalizes.

**ALEX:** And their inventory position seems to be a real competitive advantage right now. They built up $4.8 billion in inventory - 222 days - which sounds high but management is calling it an asset that lets them respond to this real-time demand environment.

**JORDAN:** One thin

This episode includes AI-generated content.

**Beta Finch Podcast Script: Qualcomm Q1 2026 Earnings**

---

**ALEX**: Welcome to Beta Finch, your AI-powered earnings breakdown where we turn complex financial reports into conversations you can actually follow. I'm Alex.

**JORDAN**: And I'm Jordan. Today we're diving into Qualcomm's first quarter 2026 results, and wow, this one's a bit of a tale of two cities.

**ALEX**: Before we jump in, I need to mention that this podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

**JORDAN**: Absolutely. So Alex, let's start with the headline numbers because they're pretty impressive on the surface.

**ALEX**: They really are. Qualcomm delivered record-breaking results - $12.3 billion in revenue and $3.50 in non-GAAP earnings per share. Both numbers hit records, with EPS coming in at the high end of their guidance range.

**JORDAN**: The breakdown is interesting too. Their chip business, QCT, hit a record $10.6 billion, while their licensing division QTL brought in $1.6 billion. But here's where it gets complicated - they're guiding down significantly for next quarter.

**ALEX**: Right, and that's the big story here. For Q2, they're forecasting total revenue of $10.2 to $11 billion, which at the midpoint represents a pretty substantial sequential decline. The handset business specifically is expected to drop from $7.8 billion to about $6 billion.

**JORDAN**: And the reason? It's all about memory shortages. CEO Cristiano Amon was very clear about this - the AI data center boom is sucking up all the high-bandwidth memory, leaving smartphone makers scrambling for DRAM.

**ALEX**: Let me read you what Amon said because it really captures the situation: "As memory suppliers redirect manufacturing capacity to HBM to meet AI data center demand, the resulting industry-wide memory shortage and price increases are likely to define the overall scale of the handset industry through the fiscal year."

**JORDAN**: It's fascinating how the AI boom is creating these ripple effects. Chinese smartphone makers in particular are being cautious, reducing their chipset inventory because they can't get enough memory to build phones.

**ALEX**: But here's what's interesting - Qualcomm is emphasizing this isn't a demand problem. Consumer appetite for premium smartphones remains strong. It's purely a supply constraint. CFO Akash Palkhiwala mentioned they saw handset units exceeding expectations in December, especially in the premium tier.

**JORDAN**: That's a crucial distinction for investors. If this were a demand issue, you'd be worried about long-term market trends. But supply constraints, while painful in the near term, typically resolve themselves.

**ALEX**: Speaking of the premium tier, Qualcomm dropped some interesting details about their market position. They're expecting about 75% share of Samsung's upcoming premium devices, which is consistent with prior expectations. And they highlighted this interesting "dual flagship" strategy where OEMs are launching multiple premium tiers.

**JORDAN**: The automotive story continues to be a bright spot. They hit another record with $1.1 billion in automotive revenue, up 15% year-over-year, and they're guiding for even stronger growth - greater than 35% year-over-year growth in Q2.

**ALEX**: The Volkswagen Group partnership announcement is huge. This isn't just about infotainment systems - Qualcomm would serve as the primary technology provider for VW's software-defined vehicle architecture, including their joint venture with Rivian. That's Audi, Porsche, the whole VW ecosystem.

**JORDAN**: And they're expanding into new territories. The robotics announcement caught my attention - they're launching a full suite of robotics technologies with the Dragon Wing IQ 10 series. They're already working with companies like

This episode includes AI-generated content.

# Beta Finch Podcast Script: NVIDIA Q4 2026 Earnings

**ALEX:** Welcome to Beta Finch, your AI-powered earnings breakdown. I'm Alex, and joining me as always is Jordan. Today we're diving into NVIDIA's absolutely massive Q4 2026 results that just dropped. Jordan, before we get started, I need to mention that this podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

**JORDAN:** Thanks Alex. And wow, where do we even start with these numbers? NVIDIA just reported Q4 revenue of $68 billion - that's up 73% year-over-year and they added $11 billion in sequential growth. This is a company that's now doing nearly $200 billion in annual data center revenue alone.

**ALEX:** Right, and what's really striking is the acceleration. They went from strong growth in Q3 to even stronger growth in Q4. The data center business hit $62 billion for the quarter, up 75% year-over-year. But Jordan, what caught my attention was their guidance for Q1 - they're calling for $78 billion in revenue, which would be another massive jump.

**JORDAN:** Exactly, and that guidance assumes zero revenue from China, which is important context given the ongoing trade restrictions. But let's talk about what's driving this growth - it's really the Blackwell architecture that's just taken off. Jensen mentioned they have 9 gigawatts of Blackwell infrastructure already deployed, and here's the kicker - even their six-year-old Ampere chips are sold out in the cloud.

**ALEX:** That supply constraint theme runs throughout this call. Colette Kress mentioned they've strategically secured inventory and purchase commitments extending into calendar 2027 - that's much further out than usual and reflects the unprecedented demand visibility they're seeing. Speaking of segments, their networking business was a real standout, hitting $11 billion in revenue, up more than 3.5x year-over-year.

**JORDAN:** And that networking growth ties directly into their "AI factory" strategy. Jensen kept emphasizing this concept that in the new world of AI, compute literally equals revenue. When companies can generate tokens faster and more efficiently, that directly translates to higher revenues. It's why their customers are so willing to spend massive amounts on infrastructure.

**ALEX:** Speaking of spending, the numbers Jensen threw out about cloud provider CapEx were staggering. He said analyst expectations for 2026 CapEx across the top five cloud providers are approaching $700 billion - that's up $120 billion just since the start of the year. But there's something bigger happening here with what they're calling "agentic AI."

**JORDAN:** Right, this was probably the most important strategic theme of the call. Jensen talked about how we've hit an inflection point with AI agents - systems like Claude Code and OpenAI Codex that can actually take on complex, long-running tasks. He mentioned these agents are being used extensively by NVIDIA's own engineers, and the demand for the compute power to run them is going exponential.

**ALEX:** And they're betting big on this trend. NVIDIA announced a $10 billion investment in Anthropic this quarter, deepening their partnerships with all the major AI players. They're also working closely with OpenAI, Meta's expanding their deployment to millions of GPUs, and they even acquired talent from Groq to enhance their inference capabilities.

**JORDAN:** Let's talk about their next-generation platform - Rubin. They unveiled this at CES with six new chips, and Jensen claims it will train models with one-fourth the number of GPUs compared to Blackwell and reduce inference costs by up to 10x. They've already started shipping samples and expect production in the second half of the year.

**ALEX:** The margins story is fascinating too. They maintained gross margins around

This episode includes AI-generated content.

**Beta Finch Podcast Script: Micron Technology Q2 2026 Earnings**

---

**ALEX:** Welcome to Beta Finch, your AI-powered earnings breakdown where we cut through the noise to bring you the market's biggest stories. I'm Alex.

**JORDAN:** And I'm Jordan. Today we're diving into Micron Technology's absolutely explosive Q2 2026 earnings that dropped yesterday. And folks, when I say explosive, I mean it – we're talking about numbers that are rewriting the record books.

**ALEX:** Before we jump in, a quick reminder that this podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

**JORDAN:** Now let's talk about these mind-blowing numbers. Alex, where do we even start?

**ALEX:** I mean, Jordan, I've covered a lot of earnings calls, but this one... Micron just posted quarterly revenue of $23.9 billion – that's up 196% year-over-year and 75% sequentially. To put that in perspective, their Q3 guidance alone exceeds the full-year revenue of every year in the company's history through 2024.

**JORDAN:** That's insane! And the margins are what really caught my eye. They're guiding for 81% gross margin in Q3. Eighty-one percent! I had to double-check that number. For context, in previous memory cycles, Micron's peak margins were in the low 60s. We're in completely uncharted territory here.

**ALEX:** Absolutely. And CEO Sanjay Mehrotra was pretty clear about what's driving this – it's all about AI. He said something that really stuck with me: "Memory makes AI smarter and more capable, enabling longer context windows, deeper reasoning chains, and multi-agent orchestration." Essentially, as AI gets more sophisticated, it becomes more memory-hungry.

**JORDAN:** Right, and what's fascinating is the supply constraint story. They're only able to fulfill 50% to two-thirds of their key customers' demand. Think about that – in a world where everyone is scrambling for AI chips, the memory bottleneck is so severe that even their biggest customers can't get what they need.

**ALEX:** Speaking of customers, let's talk about the elephant in the room – their new Strategic Customer Agreements or SCAs. They just signed their first five-year SCA, which is a big departure from their traditional one-year agreements.

**JORDAN:** Yeah, this is huge strategically. During the Q&A, analysts kept pushing for details, but Mehrotra was pretty tight-lipped about specifics due to confidentiality. What we do know is these are multi-year agreements with "specific commitments" from both sides, designed to give Micron better visibility and customers more supply assurance.

**ALEX:** And it makes sense why customers would want this. If you're NVIDIA or Microsoft planning your AI infrastructure years out, the last thing you want is to be constrained by memory availability. These SCAs essentially lock in supply, even if it means paying premium prices.

**JORDAN:** Let's talk about the HBM story because this is where things get really interesting. They're now shipping HBM4 36GB modules and have already sampled their HBM4 16-Hi product with 48GB capacity – that's a 33% increase per module. And get this – they're already working on HBM4E for 2027.

**ALEX:** The HBM ramp is incredible. Remember, high-bandwidth memory is the specialized, expensive memory that goes directly on AI accelerators. It's like the premium gasoline of the memory world, and demand is through the roof. They mentioned that AI server demand alone is driving DRAM and NAND data center bits to exceed 50% of industry TAM for the first time.

**JORDAN:** But here's what I found most interesting from the call – they're not just betting on data center AI. Mehrotra talked about on-device AI driving memory content growth everywhere. PCs with agentic AI need at least 32GB of memory, double the current average. And smartphon

This episode includes AI-generated content.

ALEX: Welcome to Beta Finch, your AI-powered earnings breakdown. I'm Alex, and I'm here with my co-host Jordan to dive into Lam Research's Q2 2026 earnings call. Before we get started, I want to remind our listeners that this podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

JORDAN: Thanks Alex. And wow, what a quarter for Lam Research! The semiconductor equipment maker absolutely crushed it with record revenues of $5.34 billion - that's their tenth consecutive quarter of growth. They beat the midpoint of guidance and exceeded expectations across the board.

ALEX: The numbers really tell the story here. For the full year 2025, they hit record revenues of $20.6 billion, up 27% year-over-year. But what's even more impressive is the profitability - gross margins of 49.9%, operating margins of 34.1%, and earnings per share of $4.89, up 49% from the prior year.

JORDAN: And looking ahead, CEO Tim Archer and CFO Doug Bettinger painted a picture of an industry that's absolutely on fire. They're projecting wafer fabrication equipment spending - that's WFE - to jump from about $110 billion in 2025 to $135 billion in 2026. That's roughly 23% growth!

ALEX: But here's the fascinating part - and this came up multiple times in the Q&A - the industry is actually constrained by clean room space shortages. Tim Archer mentioned that chipmakers are essentially sold out, which is creating this pent-up demand situation.

JORDAN: Right, it's almost like a luxury problem to have. The demand is there, the orders are there, but customers literally don't have the physical space to install all the equipment they want. Doug Bettinger mentioned they expect 2026 to be "second-half weighted" because of these constraints, with growth ramping as more clean room space comes online.

ALEX: The AI boom is really driving everything here. Lam is particularly well-positioned because AI chips require more complex manufacturing processes - specifically more deposition and etching, which is exactly what Lam specializes in. Archer talked about how they're seeing accelerated adoption of advanced technologies like gate-all-around transistors and 3D advanced packaging.

JORDAN: One product that really stood out was their Aqara conductor etch system. They've doubled its installed base over the past year and are winning production contracts for the most advanced chip manufacturing. Tim Archer explained that as critical dimensions keep shrinking - we're talking 10 to 20% smaller with each new technology node - Lam's tools become even more essential.

ALEX: The memory business is particularly interesting. DRAM was a standout, making up 23% of systems revenue, up from 16% in the previous quarter. This is largely driven by high-bandwidth memory or HBM demand for AI applications. But what caught my attention was their advanced packaging business - they expect it to grow more than 40% in 2026.

JORDAN: And then there's NAND flash memory, which has been a bit of a sleeper story. While it was down sequentially in Q2, management is seeing new use cases emerging, particularly for AI inference applications. They mentioned that for every 2-3 million AI accelerators sold, they estimate a one-point increase in NAND bit demand growth.

ALEX: Let's talk about the geographic mix because it's quite telling. China represented 35% of revenue, down from 43% in the prior quarter. Management expects China to be roughly flat year-over-year in 2026, while everywhere else grows substantially. This shift is partly due to regulatory changes but also reflects where the cutting-edge demand is coming from.

JORDAN: The guidance for Q3 2026 shows continued momentum - they're expecting revenue of $5.7 billion, plus or minus $300 million, with gross margins around 49%. What's remarkable is that they're alread

This episode includes AI-generated content.

**BETA FINCH PODCAST SCRIPT**

---

**ALEX:** Welcome to Beta Finch, your AI-powered earnings breakdown where we turn complex quarterly reports into digestible insights. I'm Alex, and I'm joined as always by my co-host Jordan. Today we're diving into KLA Corporation's Q2 2026 earnings - that's ticker KLAC for those following along.

Before we jump in, I need to share an important disclaimer: This podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

Jordan, KLA just posted some pretty impressive numbers. Walk us through the headline figures.

**JORDAN:** Alex, these results are really something. KLA delivered $3.3 billion in revenue for Q2, which represents 17% year-over-year growth. But here's what really caught my attention - their earnings per share jumped 29% to $8.85 on a non-GAAP basis. That's some serious operating leverage right there.

**ALEX:** That leverage is exactly what you want to see in a capital equipment company. And they're not just growing - they're throwing off serious cash. What did free cash flow look like?

**JORDAN:** Record-breaking, Alex. They hit $1.26 billion in quarterly free cash flow, and for the full year they generated $4.4 billion - that's 30% growth year-over-year. They returned $3 billion to shareholders through dividends and buybacks. For a company with their market cap, that's substantial capital allocation.

**ALEX:** Now, KLA is a semiconductor equipment company, specifically focused on process control - think inspection and measurement tools that ensure chips are made correctly. What's driving this growth?

**JORDAN:** It's really the AI story, Alex. CEO Rick Wallace was crystal clear that AI infrastructure demand is a core driver. Their tools are essential for manufacturing the advanced chips needed for AI applications - everything from leading-edge foundry logic to high-bandwidth memory, or HBM.

What's fascinating is they're seeing process control intensity increase dramatically, especially in memory. Wallace mentioned that DRAM manufacturing now looks "much more similar to what logic did not that long ago" in terms of requiring sophisticated inspection tools.

**ALEX:** That's a key point about intensity. Can you break that down for our listeners?

**JORDAN:** Absolutely. Process control intensity basically means how much inspection and measurement equipment you need per dollar of total semiconductor equipment spending. As chips get more complex - smaller features, more layers, tighter specifications - you need proportionally more of KLA's tools.

In DRAM memory, they're seeing about 100 basis points of intensity increase with EUV lithography adoption, and another 100 basis points with HBM. That's essentially doubling their addressable market per chip produced.

**ALEX:** Speaking of markets, let's talk guidance. What's KLA expecting for 2026?

**JORDAN:** Here's where it gets interesting. For the full year, they're guiding for mid-single digit revenue growth, but CFO Bren Higgins emphasized that growth will accelerate in the second half. They're expecting the core wafer fab equipment market to grow high single to low double digits to about $120 billion, plus an additional $12 billion advanced packaging market.

But there's a constraint story here, Alex.

**ALEX:** Right, supply constraints. This came up multiple times in the Q&A. What's the issue?

**JORDAN:** Two main problems. First, KLA themselves are constrained by long lead-time components, especially optics. Higgins said decisions they made last summer are affecting what they can ship in the first half of 2026. Their lead times are extending because demand is so strong.

But second, and this is crucial - their customers are also constrained. Multiple executives mentioned that chipmakers are "frustrated with the shells that they ha

This episode includes AI-generated content.

**ALEX**: Welcome to Beta Finch, your AI-powered earnings breakdown where we dive deep into the numbers that matter. I'm Alex.

**JORDAN**: And I'm Jordan. Today we're unpacking Intel's Q4 2025 earnings call, and wow - there's a lot to discuss here.

**ALEX**: Before we jump in, I need to mention that this podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

**JORDAN**: Absolutely. Now Alex, Intel's been on quite a journey lately, especially with new CEO Lip-Bu Tan at the helm. What were the headline numbers?

**ALEX**: The numbers actually look pretty solid on the surface. Q4 revenue came in at $13.7 billion - that's at the high end of their guidance range. They delivered non-GAAP earnings per share of 15 cents versus guidance of just 8 cents. And here's something interesting - this marks their fifth consecutive quarter of beating guidance.

**JORDAN**: That's a nice streak, but I'm sensing there's a "but" coming here, right?

**ALEX**: You know me too well, Jordan. The big story here isn't what they delivered - it's what they couldn't deliver due to supply constraints. CEO Lip-Bu Tan was pretty candid about this. He said they're "disappointed that we are not able to fully meet the demand in our markets."

**JORDAN**: Supply constraints in a chip shortage - where have we heard that before? But what's driving the demand they can't meet?

**ALEX**: It's all about AI infrastructure build-out. Their traditional server business saw 15% sequential growth - the fastest this decade, according to CFO Dave Zinsner. AI PCs were up 16% in units. But here's the kicker - Zinsner said revenue "would have been meaningfully higher if we had more supply."

**JORDAN**: So they're essentially leaving money on the table. What about their guidance for Q1 2026?

**ALEX**: This is where it gets interesting. They're guiding to $11.7 to $12.7 billion for Q1, with a midpoint of $12.2 billion. That's actually below typical seasonality. Zinsner said they'd be "well above seasonal if we had all the supply."

**JORDAN**: That's a pretty significant admission. But I noticed something in the transcript - they're prioritizing server shipments over client. Can you explain that strategy?

**ALEX**: Exactly right. They're deliberately shifting their constrained wafer supply toward higher-margin data center customers and away from PC clients. It's a smart move financially, but it shows just how tight their supply situation really is. They've basically depleted their inventory buffers and are operating hand-to-mouth.

**JORDAN**: Let's talk about their foundry ambitions. Lip-Bu has been making a lot of noise about building a world-class foundry business. What's the latest there?

**ALEX**: This is probably the most forward-looking part of the call. They're shipping products on Intel 18A - which they claim is the most advanced process manufactured on U.S. soil. But the real excitement is around Intel 14A. They expect customers to start making firm supplier decisions in the second half of 2026, with volume production targeted for 2028.

**JORDAN**: That timeline puts them roughly in line with TSMC's advanced nodes. But there was some interesting commentary about advanced packaging too, wasn't there?

**ALEX**: Yes! Zinsner said their advanced packaging opportunities - particularly something called EMIB-T - could be "well north of $1 billion" per customer engagement. He initially thought these would be "hundreds of millions" but customer interest is way stronger than expected. Some customers are even making prepayments to secure capacity.

**JORDAN**: That's a strong vote of confidence. Now, one thing that caught my attention was their ASIC business hitting a $1 billion run rate. That seems to be flying under the radar.

**ALEX**: Great point. Their custom ASIC business g

This episode includes AI-generated content.

# Beta Finch Podcast Script: Broadcom Q1 2026 Earnings

**ALEX**: Welcome to Beta Finch, your AI-powered earnings breakdown. I'm Alex, and I'm here with my co-host Jordan to dive into Broadcom's absolutely explosive Q1 2026 results. Before we jump in though, I need to mention that this podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

**JORDAN**: Thanks Alex. And wow, where do we even start with these numbers? Broadcom just delivered what might be the most jaw-dropping AI revenue guidance we've seen yet. We're talking about a company projecting over $100 billion in AI chip revenue by 2027.

**ALEX**: That's right, Jordan. Let's break down the headline numbers first. Q1 revenue hit $19.3 billion, up 29% year-over-year, crushing their guidance. But here's the kicker - they're guiding for Q2 revenue of $22 billion, which represents 47% year-over-year growth. Their AI semiconductor business alone grew 106% year-over-year to $8.4 billion in Q1.

**JORDAN**: And that acceleration is only speeding up. They're projecting AI revenue to grow 140% year-over-year in Q2 to $10.7 billion. But Alex, what really caught my attention was CEO Hock Tan's confidence about 2027. He said they have "line of sight" to achieve AI revenue from chips - just chips - in excess of $100 billion in 2027.

**ALEX**: That's an incredible statement, Jordan. And he backed it up with some pretty specific customer details. They now have six major customers for their custom AI accelerators, including a new addition - OpenAI. Let's talk about what he revealed about each customer.

**JORDAN**: Absolutely. For Google, they're continuing strong demand for seventh-generation TPUs with even stronger demand expected in 2027. Anthropic is scaling from 1 gigawatt of TPU compute in 2026 to over 3 gigawatts in 2027. And here's something interesting - Tan pushed back hard against reports that Meta's MTIA custom accelerator program was dead.

**ALEX**: Right, he was pretty emphatic about that. He said Meta's roadmap is "alive and well" and they're already shipping, with plans to scale to multiple gigawatts in 2027. Then there's the new customer, OpenAI, which is expected to deploy over 1 gigawatt of compute capacity in 2027.

**JORDAN**: What struck me most was Tan's explanation of why these partnerships are so strategic. He emphasized that for these customers, custom AI accelerators aren't optional - they're strategic necessities. These companies are competing against each other and against NVIDIA, so they need the absolute best chips, not just "good enough" ones.

**ALEX**: And that competitive advantage seems to extend beyond just the chips themselves. Broadcom is also crushing it in AI networking. In Q1, AI networking revenue grew 60% year-over-year and represented one-third of total AI revenue. In Q2, they expect that to jump to 40% of total AI revenue.

**JORDAN**: Their networking success is fascinating, Alex. They're the only company with a 100-terabit-per-second switch - the Tomahawk 6 - and they're planning to launch Tomahawk 7 in 2027 with double the performance. Tan made a great point about how they're helping customers stay on direct-attached copper instead of moving to more expensive optical solutions.

**ALEX**: Now, Jordan, I have to ask about the elephant in the room. With AI revenue growing this explosively, what about their other businesses? Their infrastructure software segment, which includes VMware, seems to be holding up well.

**JORDAN**: That's a great point. VMware revenue grew 13% year-over-year with strong bookings exceeding $9.2 billion. Tan was very clear that their infrastructure software "is not disrupted by AI." In fact, he argued that VMware Cloud Foundation is essential for enterprises running generative AI workloads.

**ALEX**: Let's talk margins

This episode includes AI-generated content.

ALEX: Welcome to Beta Finch, your AI-powered earnings breakdown. I'm Alex, and joining me as always is Jordan. Today we're diving into Broadcom's Q3 2024 earnings, and wow - this was quite the quarter for CEO Hock Tan and his team.

Before we jump in, I need to share our standard disclaimer: This podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

So Jordan, let's start with the headline numbers - what jumped out at you?

JORDAN: Alex, these numbers are pretty remarkable. Broadcom posted $13.1 billion in revenue, up 47% year-over-year. Operating profit was up 44%. But here's the thing - when you strip out VMware, which they acquired last year, the underlying semiconductor business only grew 4% year-over-year. So this is really a story of two businesses firing on different cylinders.

ALEX: Right, and speaking of VMware, that integration seems to be going better than expected. Can you break down what's happening there?

JORDAN: Absolutely. VMware contributed $3.8 billion in revenue this quarter, and Hock Tan was pretty excited about their transformation strategy. They're aggressively moving customers from perpetual licenses to subscription models, specifically pushing something called VMware Cloud Foundation - that's their full virtualization stack.

What's impressive is they booked over 15 million CPU cores of this product, representing 80% of total VMware bookings. That translated to $2.5 billion in annualized booking value, up 32% from the prior quarter.

ALEX: And they're cutting costs at the same time, right?

JORDAN: Exactly - classic Broadcom playbook. They brought VMware's operating expenses down to $1.3 billion from $1.6 billion in Q2. Tan said they're on track to hit their target of $8.5 billion in adjusted EBITDA within three years, and maybe even exceed it by fiscal 2025.

ALEX: Now let's talk about the elephant in the room - AI. Everyone wants to know how Broadcom is riding this wave.

JORDAN: AI is absolutely driving their semiconductor business. Tan said AI revenue will hit $3.5 billion in Q4, bringing the full year to $12 billion - up from their previous guidance of $11 billion. That's huge growth.

The interesting part is the mix - about two-thirds is custom AI accelerators and one-third is AI networking. These are the chips that power AI data centers for the big hyperscalers like Google, Amazon, Microsoft.

ALEX: There was some interesting commentary about custom chips versus off-the-shelf GPUs. What's Tan's view on where this market is headed?

JORDAN: This was fascinating, Alex. Tan actually said he's changed his mind on this. He used to think general-purpose merchant chips would win - that's typically how the semiconductor industry works. But now he believes the big cloud companies have the scale and financial resources to justify building their own custom AI accelerators.

His logic is compelling: if GPUs are more important than engineers to these companies right now - and he literally said that - then controlling your own silicon destiny makes sense. He sees this trend accelerating, though it'll take time.

ALEX: What about the non-AI semiconductor business? That's been struggling, right?

JORDAN: Yeah, but here's the key - Tan believes they've hit bottom. Non-AI bookings were up 20% year-over-year in Q3, which is a strong leading indicator. Networking revenue outside of AI was up 17% sequentially, even though it was still down 41% year-over-year.

The company sees different recovery timelines by segment - server and storage are showing early signs of improvement, wireless should get a boost from new device launches, but broadband remains weak due to telecom spending cuts.

ALEX: There were some good questions from analysts during the Q&A. Anything particularly noteworthy?

JORDAN: One analyst asked about th

This episode includes AI-generated content.

**Beta Finch Podcast Script**

ALEX: Welcome to Beta Finch, your AI-powered earnings breakdown. I'm Alex, and I'm here with my co-host Jordan to dive into some fascinating quarterly results. Jordan, we've got Amphenol's Q4 2025 earnings to discuss today - and wow, what a quarter this was.

JORDAN: Absolutely, Alex. But before we get into these impressive numbers, I need to share our standard disclaimer. This podcast is AI-generated content for educational and entertainment purposes only. Nothing we discuss should be considered investment advice. Always do your own research and consult a qualified financial advisor before making any investment decisions.

ALEX: Thanks Jordan. Now, let's talk Amphenol - because these results are genuinely remarkable. We're looking at a company that just posted record sales of $6.4 billion for the quarter, up 49% in US dollars and 37% organically. For the full year, they hit $23.1 billion in sales. Jordan, they've literally more than doubled their revenue in just four years.

JORDAN: That's incredible scale, Alex. And what really caught my attention was their book-to-bill ratio - 1.31 to one in Q4, driven by a record $8.4 billion in orders. That's 68% growth in orders compared to last year. CEO Adam Norwitt was pretty clear about what's driving this: AI infrastructure investments are creating unprecedented demand for their interconnect products.

ALEX: Right, and this isn't just a revenue story. Their adjusted operating margin hit 27.5% in the quarter - that's a 510 basis point improvement year-over-year. For a company growing this fast, those margin numbers are exceptional. Jordan, what did you make of their strategy around acquisitions?

JORDAN: This is where it gets really interesting, Alex. They just closed their largest-ever acquisition - the CommScope CCS business - for what became over $4 billion in annualized sales after strong momentum. Norwitt was fascinating when he talked about this deal. He said they don't use the words "integration" or "synergy" at Amphenol. Instead, they let acquired companies evolve into the Amphenol family while maintaining their entrepreneurial culture.

ALEX: That's such a unique approach. And strategically, this CommScope acquisition is huge for them. It dramatically expands their fiber optic capabilities, which complements their traditional strength in high-speed copper interconnects. As Norwitt put it in the call, they can now offer customers solutions across "the entirety of the interconnect spectrum."

JORDAN: Exactly. And the timing couldn't be better with AI driving demand for both copper and fiber solutions. Speaking of AI, let's break down their IT datacom segment - it represented 38% of sales in Q4 and grew 110% organically. That's not a typo, folks - one hundred and ten percent growth.

ALEX: The breadth of their AI business really stood out to me, Jordan. Norwitt emphasized they don't have any 10% customers, meaning they're diversified across the entire AI stack - from hyperscalers to equipment manufacturers to chip designers. That's a much safer position than being dependent on one or two major customers.

JORDAN: And those record orders we mentioned? They're giving customers extended order windows to help share investment risk for these complex, high-tech products. Customers are essentially making solid commitments that give Amphenol confidence to make the capital investments needed for these next-generation systems.

ALEX: Let's talk about what this means geographically. One surprise was their strength in Europe, where they saw robust organic growth in both automotive and industrial markets. CFO Craig Lampo noted their strongest automotive growth in Q4 was actually in Europe - quite different from the doom and gloom we usually hear about that region.

JORDAN: That's a great point. And looking ahead, their Q1 2026 guidance is impressive: $6.9 to $7 billion in sales and $0.91 to $0.93 in adjusted EPS. That represents 43-45% sales growth and 44-48%

This episode includes AI-generated content.