Broadcom, Intel CEO Warn of Capital as a Bottleneck for AI Expansion; Multiple Semiconductor Stocks in U.S. Market Fall
The Block
1h ago
Ai Focus
On October 12th in the US stock market, several AI semiconductor industry chain stocks declined. Broadcom and Intel CEO stated at a DBS Asia Future Forward Conference held in Singapore that financing may become a greater obstacle to AI expansion than semiconductor, memory chip, and power shortages. Both also believe that the demand for AI infrastructure will remain strong, and the imbalance between supply and demand in the industry may continue until 2029 or even 2030.
Helpful
No.Help

On October 12th in the US stock market, several stocks in the AI semiconductor industry chain declined. NVIDIA (NVDA.O) fell by 1.29%, Broadcom (AVGO.O) fell by 1.88%, ASML (ASML.O) fell by 1.69%, AMD (AMD.O) fell by 2.28%, and ARM (ARM.O) fell by 4.15%. Among storage stocks, SanDisk (SNDK.O) fell by 5.97%, Micron Technology (MU.O) fell by 3.88%, Seagate Technology (STX.O) fell by 3.8%, Western Digital (WDC.O) fell by 2.6%, and SK Hynix (SKHY.O) fell by 4.02%.

On the news front, during the DBS Asia Future Forward Conference held in Singapore on Monday local time, Broadcom and Intel CEO expressed their views regarding the bottlenecks in the development of AI. Both mentioned that financing could become a major constraint for the expansion of AI.

Broadcom's CEO Chen Fuyang stated that as the industry competes to build the infrastructure necessary to power models, financing may become a greater obstacle to AI expansion than semiconductor shortages, memory chips, or power shortages. Capital could be the biggest limiting factor.

Intel's CEO Chen Liwu mentioned that the capital cost of AI company is rising, and people are also questioning whether companies that make large investments in AI can generate sufficient returns. 'If there is any challenge ahead, it is capital, and the longer-term issue is that people hope to see a return on investment.'

Previously, the industry was more concerned with whether the production of hardware devices in the AI industrial chain could support the construction of AI infrastructure. Chen Fuyang stated that over the past two years, the industry has repeatedly encountered supply bottlenecks, but with the expansion of production capacity, some of these constraints will be alleviated. For example, the current storage shortage may no longer exist in a year, and the wafer shortage could be resolved by 2028 or 2029. Power supply will also improve with the deployment of new infrastructure.

Not only Broadcom and Intel, but NVIDIA has also recognized the financing needs of AI manufacturers and has become involved in the financing aspect of the AI industry chain. In August, NVIDIA announced strategic partnerships with Apollo, BlackRock, Fortress Investment, Goldman Sachs, and KKR to establish an independent financing platform, mobilizing over $500 billion in third-party capital to support the construction of AI infrastructure.

Recently, there have been reports that Botong is in preliminary discussions regarding a $50 billion debt financing deal to help OpenAI raise funds for the development and procurement of custom chips.

The deep equity ties and cooperative relationships between multiple semiconductor companies and AI manufacturers are the backdrop to semiconductor firms' concerns about whether the financing capabilities of AI manufacturers can support infrastructure construction. Broadcom collaborates with OpenAI, Google, and several other AI manufacturers to develop customized AI chips; AMD is a supplier of chips and rack-level systems for multiple AI manufacturers; NVIDIA is a shareholder and computing power provider for OpenAI.

Even as the industry faces financial challenges, Broadcom and Intel CEO remain optimistic about the future demand for AI infrastructure. Chen Fuyang stated that the demand for computing infrastructure has exceeded the company's supply capacity, and this imbalance between supply and demand may continue until 2029 or even 2030. Chen Liwu, on the other hand, believes that with the widespread adoption of AI applications and the growing demand for computing power, it is unlikely that there will be a significant slowdown in industry development in the next two to three years.

Tip
$0
Like
0
Save
0
Views 20
HKWDB reminds readers to view blockchain rationally, stay aware of risks, and beware of virtual token issuance and speculation. All content on this site represents market information or related viewpoints only and does not constitute any form of investment advice. If you find sensitive content, please click“Report”,and we will handle it promptly。
Submit
Comment 0
Hot
Latest
No comments yet. Be the first!
Related
Microsoft Announces Its Largest Self-Developed Chip Bet Ever; Market Value Returns to $4 Trillion
Microsoft Chairman and CEO Satya Nadella stated that the company announced its largest bet to date on self-developed chips. Microsoft claims that Azure Maia 200 has entered production, and Azure Maia 300 has been released, with plans to expand the Maia series to several gigawatts of deployment starting from 2027. Influenced by this news, Microsoft's stock price rose by 1.14% on that day, bringing its market value back to $4 trillion.
The Block
·2026-10-13 12:12:31
9
XRPN vs XRP: How will financial packaging carriers trade with the tokens themselves?
The merger of Evernorth with Armada Acquisition Corp and II has been completed. XRPN began trading on NASDAQ on October 12th, holding approximately 473 million XRP on its balance sheet. Reports indicate that this provides investors with a real-time comparison of how the performance of the listed company, which holds these XRP, as a “packaging” vehicle, will compare to that of the XRP tokens themselves; and this listing occurred in a market environment that was more cautious than during the first wave of enthusiasm for financial companies.
Coinpedia
·2026-10-13 12:03:09
12
MONTECH Launches CyberFlow Liquid-Cooled 32mm Thick Array, Magnetic吸附 VRM Fan
MONTECH announced yesterday the launch of the CyberFlow series of processors and AIO liquid cooling radiators, which are available in two versions: Core and ARGB. Both versions are equipped with 32mm thick fans and magnetic mounting. This series is compatible with Intel's LGA 1700/1851 as well as AMD AM5 / AM4 platforms, with a 360mm design. The Core version is priced at NT$2,590, and the ARGB version is priced at NT$2,790.
The Block
·2026-10-13 11:31:35
23
NVIDIA unveils Vera Rubin rack design to promote unified standards in AI factories
NVIDIA announced the opening of Vera Rubin rack and computing bracket designs at the OCP Global Summit, and jointly released AI infrastructure open standards with multiple technology companies, covering aspects such as power supply, cooling, networking, and grid resilience.
The Block
·2026-10-13 11:21:39
18
View More