Why I Keep Buying AMAT Before Aug. 13: AI's Real Bottleneck Is Vertical Stacking

Why AMAT still looks interesting into earnings

I keep buying AMAT before next week's earnings because the setup is straightforward. Applied reports Aug. 13, 2026 at 4:30 p.m. ET, and the last quarter already cleared a basic credibility check. In Q2, Applied posted $2.86 EPS versus a $2.68 estimate, a 6.72% beat. After that report, the shares kept climbing 22.5% higher, even though the stock only moved −0.9% the day after earnings. That tells me the market was interested, not fully convinced.

The core bull case is simple: AI chips are increasingly constrained by memory, and Applied has equipment aimed at the 3D-stacking workflows that HBM and advanced packaging require. Its recently unveiled advanced packaging portfolio covers several of those steps, so if vertical stacking becomes a bigger part of AI build-outs, Applied is clearly inside that path.

That is why I am constructive into fiscal third-quarter results. I am not arguing one strong quarter proves a multi-year trend. I am arguing that if Applied shows even a portion of that same execution again, the stock may still have room to work.

Applied's packaging pitch rests on a real manufacturing problem

The key question is not whether AI spending exists. It is whether Applied is solving a real bottleneck or just selling expensive equipment language.

Higher-yield stacking is the actual customer need

Applied's own release frames the issue clearly. AI compute is increasingly limited by memory, and that is accelerating adoption of advanced packaging architectures, including high bandwidth memory (HBM) and 3D stacking. The company says its new packaging systems target the most critical advanced packaging steps, delivering higher-yield chip stacking for HBM and logic.

That matters because 3D builds only work if the critical interconnect steps are reliable. Applied is not pitching a niche novelty; it is pitching tools for the steps that can make or break yield in stacked-memory architectures.

A smaller footprint can be as important as raw performance

Applied also says the new epitaxy system adds a critical logic-class step-boosting memory speed and efficiency while maximizing output within tight fab footprint and supply constraints.

That is useful for two reasons. First, it brings logic-style process capability into DRAM manufacturing. Second, if customers can get more output from the same footprint, that can matter as much as a spec sheet in a capacity-constrained world.

Other vendors are seeing the same packaging traffic

KLA's commentary gives an outside view on the same trend. KLA says advanced packaging demand is strong enough that its advanced-packaging process-control revenue should reach about $1.1 billion in calendar 2026, with more than 70% year-over-year growth.

That does not prove Applied is winning every relevant battle. But it does suggest the packaging shift is real enough that suppliers across the equipment stack are starting to feel it.

What Aug. 13 needs to confirm

What gets checked on Aug. 13 at 4:30 p.m. ET is whether vertical stacking is moving from a product story toward real customer urgency. I would rather hear concrete demand signals than another broad semiconductor pep talk.

What to watch on the call

I would focus management answers on a short list:

• Whether packaging and HBM-related tool demand is becoming a larger part of customer roadmaps.
• Whether customers are treating these tools as necessary steps rather than optional enhancements.
• Whether the new DRAM and packaging systems are showing up in customer discussions as capacity-building spends.

If management talks up packaging but the broader memory story still sounds weak, the thesis needs more proof.

Why the post-earnings tape matters

The price action after the last report is worth keeping in mind. AMAT shares went down −0.9% the day following the earnings announcement, but then drifted 22.5% higher over the next stretch.

That does not guarantee the same outcome this time. It does suggest the market wanted more confirmation, not just a headline beat. If the call and the tape agree after Aug. 13, the bull case stays intact. If they do not, the recent run may have gotten ahead of the evidence.