"The Chip Duopoly" Reshapes Trading Logic with a Frenzied Surge: Under the Pressure of the AI Bull Market, Equity Dilution is Negligible

In the global stock market, issuing new shares or high-level placements is often seen by investors as the "strongest formula" for a shift from bullish to bearish sentiment, as shareholder equity gets diluted by new shares. However, in this unprecedented global frenzy of artificial intelligence investment, this traditional logic has been completely overturned, with the chip giants—Intel and AMD—experiencing significant stock price increases despite the dilution and thinning environment.

First, let's take a look at the American tech giant Intel Corporation (INTC.US). Since early August, this continuously loss-making chip manufacturer has initiated a capital market action to sell a total of $18 billion worth of shares to the U.S. government, SoftBank Group Corp, and its long-time competitor NVIDIA, resulting in its stock price soaring nearly 90%. This highlights the global capital's fervent pursuit of chip giants like NVIDIA, AMD, and Intel amid the AI investment frenzy. These transactions aim to strengthen Intel's weak balance sheet with fresh capital.

According to institutional estimates, these new shares will "dilute" the equity value of existing Intel shareholders by about 14%. If the warrants held by the U.S. government are exercised under certain conditions, Intel shareholders may face even greater dilution. However, for now, investors are far more enthusiastic about the "significant breathing space" represented by this government cash than they are concerned about the dilution of individual share value.

Recently, the prices of high-performance DRAM and NAND series storage products have surged, coupled with the fact that the world's highest-valued AI startup, OpenAI, has secured over $1 trillion in AI computing infrastructure deals, and "chip foundry king" TSMC has reported exceptionally strong earnings that exceeded expectations and raised its revenue growth forecast for 2025 to the mid-30% range. Together, these factors have significantly reinforced the "long-term bull market narrative logic" for AI GPU, ASIC, HBM, data center SSD storage systems, liquid cooling systems, core power equipment, and other AI computing infrastructure sectors.

The demand for AI computing power driven by generative AI applications and AI agents at the inference end is akin to "stars and the sea," expected to drive the artificial intelligence computing infrastructure market to continue exhibiting exponential growth. The "AI inference system" is also seen by Jensen Huang as the largest source of future revenue for NVIDIA.

According to Wall Street financial giants Morgan Stanley, Citigroup, Loop Capital, and Wedbush, the global investment wave in artificial intelligence infrastructure centered around AI computing hardware is far from over; it is merely at the beginning. Under the unprecedented "AI computing demand storm," this round of AI investment is expected to reach a scale of $2 trillion to $3 trillion.

As the earnings season for the stock market opens, the impressive performance of TSMC and ASML undoubtedly significantly reinforces the "long-term AI bull market" narrative logic in the global stock market and the "AI investment faith" of tech stock believers. This also means that the current "super bull market" led by NVIDIA, TSMC, Broadcom, and Micron Technology in the global AI computing industry chain is far from over, and this industry chain will remain the most favored investment sector for global capital in the foreseeable future Equity Transfer and Warrant Trading Ignite Valuation of Chip Giants

"Investors have priced in all these existing or potential positives in advance, significantly downplaying the real actual costs—there is a huge dilution space accompanying all these share sales," said Michael Bailey, research director at Fulton Breakefield Broenniman, which manages $2.3 billion in assets. "This is a situation where both pros and cons coexist. However, investors are betting that the pros outweigh the cons."

As Intel investors face dilution in a wave of equity sales, about 14 cents of every dollar of profit will go to SoftBank, the U.S. government, and shares held by NVIDIA.

The same is true for AMD (AMD.US), a long-time competitor of NVIDIA and Intel. The company signed an agreement with AI application leader OpenAI on October 6, which could soon bring in billions of dollars in revenue. In exchange, AMD granted the ChatGPT operator equity subscription rights that vest upon reaching specific stock price milestones. Although AMD's earnings per share will be lower if OpenAI's bet succeeds, AMD's stock price still soared 43% in the three trading days following the announcement.

The math behind equity dilution is simple: buying a company's stock essentially equates to obtaining a "theoretical slice" of its profits, reflected in earnings per share (EPS). When new shares are issued, profits do not change, but the portion allocated to each share becomes smaller. This is in stark contrast to stock buybacks, which typically drive stock prices significantly higher, as investors are entitled to claim a larger share of the profit when the number of outstanding shares decreases.

Of course, selling shares will bring in a large amount of cash in the short term—this is almost entirely the intention of Intel's management. The company is in the midst of an expensive chip factory expansion, trying to transform into a contract chip manufacturing giant similar to TSMC (TSM.US).

The problem is that Intel itself may need to do more, as the capital it is raising is only a little more than half of the approximately $30 billion needed to build a large chip manufacturing plant. This may also explain why the company reportedly approached one of its clients, Apple Inc., for investment at the end of last month. This report drove Intel's stock price up 21% in three trading days.

For investors, the bullish logic on Intel mainly lies in the strong cash flow brought by the U.S. government and NVIDIA's $5 billion investment in Intel, which together with its long-time rival Intel, is poised to reshape the development landscape of PCs and data centers. Especially with NVIDIA planning to increase its layout in AI PCs, it is expected to leverage the most powerful x86 architecture ecosystem on the PC side to fully penetrate the high-growth "edge AI field" of AI PCs, potentially ushering in a new revenue increment for NVIDIA, while Intel is expected to leverage NVIDIA to reach the very center of the AI core track In terms of NVIDIA's current core revenue-generating business—data center (DC) business, NVIDIA NVLink has successfully integrated x86 architecture + GPU, launching a new artificial intelligence computing power integration system of "custom x86 architecture CPU + NVIDIA AI GPU + CUDA acceleration platform." The collaboration of "CPU + GPU" will undoubtedly give NVIDIA more say in the architectural construction of cloud computing vendors/supercomputing/enterprise DC.

The two architectures will be "seamlessly connected," with Intel creating a custom x86 CPU for NVIDIA, integrated into NVIDIA's AI computing infrastructure platform. The core is to utilize the bandwidth/interface advantages of NVLink and architectural consistency for high-speed coupling of x86 architecture CPU + NVIDIA Blackwell/future Rubin AI GPU, targeting system-level bottlenecks in large model training/inference, focusing on a more complete "system-level/AI computing cluster platform" solution. Intel will leverage the x86 architecture (non-foundry) that occupies the core architectural ecosystem of data centers to enter the AI infrastructure "golden growth zone."

Better to have than not to have

"For shareholders, I think the calculation is: would you rather own 80% or 70% of 'something,' or nothing at all?" said Jay Goldberg, a senior analyst at Seaport Global, who recently upgraded Intel's stock rating from "sell" to "neutral." "There are a lot of concerns about how Intel will survive; they urgently need funding and new sales channels, and investors are willing to accept this dilution because it is a path to save the company."

On the other hand, the massive financing around AI may also mask potentially more severe issues within companies. For example, Intel needs to attract a customer base for its foundry business; before this happens, any rise driven by stock sales or deep cooperation with NVIDIA is "unsustainable," wrote HSBC stock analyst Frank Lee in a report last week that downgraded Intel's rating to equivalent to "sell."

In contrast to Intel, AMD's situation is different. This chip giant has seen rapid performance growth in recent years—its net profit in the most recent quarter more than doubled to $872 million, with revenue soaring 32% to $7.7 billion. However, compared to industry leader NVIDIA, the company holds only a small share of the AI GPU market used for massive AI training/inference workloads.

HSBC reiterated its "buy" rating on AMD (AMD.US), a long-time competitor of NVIDIA, in its latest bullish research report, significantly raising its target price from $185 to $310, the highest target price level on Wall Street. Analyst Lee wrote that the company's recent deal with OpenAI "presents a more favorable bullish picture" and enhances performance forecast clarity and market bullish sentiment; he also added, "There may be further upside from pricing premiums and additional AI GPU shipment expansion." As of Tuesday's close, AMD's stock price has risen 35% since October, with an increase of over 80% year-to-date Therefore, this deal with OpenAI is a significant victory for AMD's data center AI GPU business. However, the warrants contain an unusual "twist." If specific financial and stock price milestones are reached, OpenAI will have the right to purchase up to 160 million shares of AMD stock at a price of one cent per share, accounting for about 10% of its outstanding shares. These warrants will vest in stages, with the vesting conditions related to the infrastructure based on AMD AI GPU deployments. AMD has redacted other details regarding the conditions attached to each stage for the media, with the only disclosed detail being the final vesting of warrants when the chip giant's stock price reaches $600. At this milestone stock price level, the company's market capitalization will reach $1 trillion.

This deal was reached a few weeks after NVIDIA agreed to invest up to $100 billion in OpenAI. OpenAI is undertaking several large-scale AI data center constructions, utilizing not only NVIDIA AI GPU computing clusters but also AI ASIC clusters developed in collaboration with Broadcom and AMD AI GPUs. However, OpenAI has yet to achieve profitability, and such financing arrangements evoke typical characteristics of the internet bubble era: mutually beneficial transactions between loss-making companies make performance growth appear stronger than it actually is.

"There is this unusual 'circular financing' happening here, as these tech companies are looking for creative ways to achieve stronger growth," said Bailey from Fulton Breakefield Broenniman. "Investors are doing the math, and they currently feel comfortable with this arrangement. But if we look back at Wall Street history, this is unusual."