RSS Amplifier

Key Context by Tae Kim · Jul 14, 2026

Why the AI Trade Is Unstoppable

0
Sign in to vote or save

Tae Kim · Key Context by Tae Kim

The bears are talking again about the unsustainability of the AI infrastructure trade. Like the last dozen times over the past three years, they will likely be wrong again.

Two things are driving stocks lower this month: 1) geopolitical uncertainty and 2) hand-wringing about the future fundamentals of the AI trade.

Yes, unfortunately, President Trump has decided to escalate the war in Iran again. I don’t know how long this new round of kinetic bombing will last or when it will end. But history has shown us that as soon as the market starts dropping significantly, he usually backs down. Given that track record, we should expect the same this time.

Back to the underlying fundamentals. When stocks fall, I’ve have noticed a tendency for some people to lose their nerve and start guessing or manufacturing narratives that aren’t backed by real new evidence or accurate information.

Has anything materially changed? I don’t believe so. I haven’t seen any negative fundamental evidence yet. In fact, as with every prior pullback in AI stocks, the fundamentals seem to be IMPROVING.

Here are the latest round of 7 positive data points and why I believe the skeptics are misunderstanding reality below:

  1. Meta FUD. Negative speculation about Meta’s future AI capex plans was the first thing that drove AI stocks lower at the beginning of the month. But new reporting now shows Meta is likely to INCREASE its AI data center spending, not decrease it.


    Reuters on July 9: Meta “plans to double capacity again next year to reach ​a total of 14 gigawatts in 2027.”


    CNBC on July 13: Meta “said in a blog post on Monday that the site in Richland Parish, Louisiana, home to what will be Meta’s largest data center, will be a 5GW facility and cost over $50 billion. That’s higher than the $27 billion figure that was revealed in October”

  2. Foxconn, a large maker of AI servers, reported its biggest year-over-year revenue growth this year, an acceleration for the month of June.

    3. King Slide, an Nvidia AI server rail supplier, posted accelerating revenue growth of 220.5% year-over-year for June.

    4. SK Hynix is the largest maker of HBM memory for AI servers. Last Friday, SK Chairman Chey Tae-won said on Bloomberg TV that customers have asked for 5x–6x more capacity, significantly more than the 2x capacity SK Hynix had promised for the next five years. He said on CNBC that when he told a customer SK Hynix planned to double capacity over the next five years, the customer replied, “THAT’S NOT ENOUGH MAN. WE NEED MORE” That person sure sounds a lot like Jensen.

    5. Quotes from Morgan Stanley last week based on investor meetings with Nvidia CEO Jensen Huang, CFO Colette Kress, and head of IR Toshiya Hari:

Read the original on taekim.substack.com

Comments

Nothing yet. Say the first thing.

    Sign in to join the conversation.