Chip Selloff Deepens as Investors Question AI Trade
Read the Portuguese translation →
Author: Bloomberg Television · Watch on YouTube
Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.
Original language: English
Original video: Chip Selloff Deepens as Investors Question AI Trade · Channel: Bloomberg Television · 01:41 · Watch on YouTube
Topics: Task and workflow automation, AI infrastructure and hardware, Markets and economy
Summary
Section titled “Summary”A selloff in chip stocks sent South Korea’s KOSPI down 7% as concerns about the AI trade resurfaced. Broadcom and Intel executives questioned how businesses will profit from AI and warned that spending on experimentation could become a capital problem. Matt Bloxham says experimentation is necessary, while businesses are still developing appropriate guardrails for AI use. He sees financing as less of a near-term constraint than data-center approvals, hardware availability, and other practical challenges, though financing could become an issue later if AI’s productivity benefits remain uncertain.
Key points
Section titled “Key points”- Chip stocks sold off, and South Korea’s KOSPI fell 7%.
- Broadcom’s CEO compared market sentiment to being in a casino and said financing could be a major problem.
- Broadcom and Intel executives said it remains unclear how businesses will profit from AI.
- Some businesses are spending on AI tokens and experimenting without clear returns, raising concerns about wasted capital.
- Bloxham says many companies are still experimenting and have not necessarily established enough guardrails for employee AI use.
- He views experimentation as necessary for businesses to understand how AI can help them.
- Bloxham says hundreds of billions of dollars have been raised for the AI trade through equity and debt markets this year.
- In the near term, data-center approvals and access to the hardware needed to expand capacity may be bigger constraints than financing.
- Financing could become a more significant issue if doubts grow about AI’s productivity benefits.
Chapters
Section titled “Chapters”| Start | Chapter | Summary |
|---|---|---|
| 00:00 | Chip selloff and renewed AI concerns | The KOSPI falls 7% as executives raise questions about AI profitability, spending, and financing. |
| 00:33 | AI experimentation and business guardrails | Bloxham says companies are still experimenting with AI and learning how to use it, while developing guardrails. |
| 01:09 | Near-term capacity constraints and longer-term financing | Bloxham points to data-center approvals and hardware as near-term challenges, with financing risks potentially emerging later if AI productivity remains uncertain. |
Quotes
Section titled “Quotes”A sell off in chip stocks sending Korea’s KOSPI down 7%. — 00:00
how businesses profit from AI is still a question mark. — 00:00
you need to experiment to understand how AI can help your business. — 00:33
there’s definitely plenty of capital in the system to support the builds out. — 01:09
the financing issue could come beyond that if there are more question marks about the productivity benefits of AI. — 01:41
Concepts and entities
Section titled “Concepts and entities”Broadcom, Intel, Matt Bloxham, KOSPI, AI trade, tokens, data centers
Full transcript (original language)
Section titled “Full transcript (original language)”00:00 A sell off in chip stocks sending Korea’s KOSPI down 7%. AI fears creeping back up as Broadcom’s CEO compared market sentiment to being in a casino, saying financing will be a big problem. Joining us now is Matt Bloxham of Bloomberg Intelligence. And it wasn’t just Broadcom CEO either, Matt. It was also Intel’s. They were on a panel together. And between them, they said things like how businesses profit from AI is still a question mark. A lot of businesses are spending on tokens and just basically messing around. They’re wasting money, and that capital could be a major issue. How seriously should we take these comments in the sense
00:33 of maybe getting out of some of these stocks? Yeah. Well, look. You know, I think they’re all kind of quite serious issues and relevant issues. You know, I think it’s certainly true that a lot of businesses right now are still experimenting with AI and, you know, haven’t necessarily put too many kind of guardrails around how their employees use them. You know, we’re starting to see some signs of that, but, you know, you need to experiment to understand how AI can help your business. It’s gonna be a kind of evolution… Evil… Evolving issue. You know, I think that the financing potentially
01:09 kind of a longer term thing personally. You know, if you look at this year, you know, hundreds of billions of dollars have been raised, for the AI trade across both equity and debt markets. So I think at least for the the near term, maybe 2027 and into 2028, there’s definitely plenty of capital in the system to support the builds out. I would say in the near term, probably what’s a bigger issue for kind of capacity expansion is the kind of the logistics and practicalities of getting approvals for data centers, getting the actual hardware that you need to kind of build that capacity. I think the financing issue
01:41 could come beyond that if there are more question marks about the productivity benefits of AI.