Nvidia's support for CoreWeave to build a &#34;computing-finance&#34; closed loop is simply too enticing, forcing Google and AMD to &#34;feel their way across the river,&#34; attempting to invest real money to create their own AI chip ecosystem.
On February 20, The Wall Street Journal cited sources saying that Google is exploring the use of its substantial financial resources to build a broader AI ecosystem, with the core strategy being to provide financial support to data center partners, on the condition that they must use Google's TPU chips. Meanwhile, Wall Street Insight wrote that AMD is also aggressively promoting its AI chips by providing guarantees for loans to customers.
This strategy is seen by the market as a direct replication of Nvidia's &#34;CoreWeave model&#34;—that is, by supporting &#34;emerging cloud&#34; vendors, bypassing traditional cloud giants that have already been occupied by Nvidia or have self-developed chips, to establish its own &#34;direct forces.&#34;
<h2>Imitating Nvidia's Route: Investing in &#34;New Cloud&#34; and Mining Companies</h2>
Sources revealed that Google is in talks to invest about $100 million in the cloud computing startup Fluidstack. This deal values Fluidstack at approximately $7.5 billion.
This is not just a financial investment but also a strategic binding. Companies like Fluidstack, classified as &#34;emerging cloud&#34; firms, are important variables in the current AI computing power market, as they specialize in providing computing power services for AI companies. Previously, Nvidia supported CoreWeave, allowing it to stockpile GPUs significantly, thus creating a breakthrough in the market.
Now, Google wants to &#34;copy the model.&#34; Sources say that Google hopes to amplify Fluidstack's growth potential through investment and encourage more computing providers to use its AI chips.
In addition to direct investments, Google has also extended its reach to former cryptocurrency mining companies that are undergoing transformation.
Reports indicate that Google has provided financing support (backstopped financing) for projects involving companies like Hut 8, Cipher Mining, and TeraWulf. These companies have ready-made data center infrastructure and are eager to transform into AI computing power factories, while the funding support from Google will lead to their acceptance of TPUs.
<h2>AMD's Aggressive &#34;Bet&#34;: If It Can't Sell, I'll Rent</h2>
Compared to Google's direct investment, AMD's approach appears more aggressive and risk-tolerant.
According to reports, AMD will provide substantial guarantees for a $300 million loan to the data center startup Crusoe. This funding from Goldman Sachs will be used to purchase AMD's AI chips.
The most notable aspect is the &#34;backstop clause.&#34; Sources revealed that if Crusoe cannot find customers to use these chips, AMD agrees to rent these chips from Crusoe. This means AMD becomes the &#34;last renter,&#34; alleviating concerns on the demand side for customers
Although this approach can temporarily boost sales, it also exposes chip manufacturers to significant risks when AI demand slows down. Once the market cools, AMD's balance sheet will directly bear the impact.
<h2>Why Take the Long Way Around?</h2>
The reason Google and AMD have chosen this &#34;detour&#34; is that the main road has been blocked.
For Google, although star startups like Anthropic are expanding the use of TPUs, traditional cloud competitors (such as Amazon AWS and Microsoft Azure) show little interest in TPUs.
Industry insiders point out: &#34;The interest of major cloud service providers seems lukewarm, partly because they view Google as a competitor.&#34; Additionally, Amazon AWS is also heavily developing its own AI chips.
Therefore, supporting neutral third-party &#34;emerging clouds&#34; has become the best choice for Google and AMD to break the blockade.
<h2>Internal Power Struggles and Capacity Bottlenecks</h2>
To accelerate the commercialization of TPUs, Google has even considered structural adjustments internally.
According to insiders, some management in Google's cloud division recently revisited a long-standing internal debate: &#34;Should the TPU team be reorganized into an independent department?&#34; This split plan could allow Google to bring in external capital to expand investment opportunities.
However, this suggestion has been denied by Google officials. A Google representative clearly stated that &#34;there are no plans to reorganize the TPU department&#34; and emphasized that &#34;maintaining close integration of the chip team with other departments in the company (such as the Gemini model development team) is advantageous.&#34;
Beyond structural issues, a more realistic roadblock is capacity.
Although Alphabet has currently partnered with Broadcom to design TPUs, with TSMC as the foundry, TSMC's advanced capacity is stretched thin amid the global surge in AI demand.
Insiders say, &#34;TSMC may prioritize its largest customer, NVIDIA, over Google.&#34; Additionally, the global shortage of HBM memory chips, which are essential for AI chips, is also a significant challenge that Google must face in expanding TPU shipments

GOOW

SOXQ

Google is trying to replicate Nvidia's successful path by leveraging its financial strength to build a chip ecosystem. The company is in talks to invest about $100 million in the "new cloud" company Fluidstack and is providing project financing support to transitioning mining companies like Hut 8 in exchange for the adoption of TPU chips. AMD is even more aggressive, providing guarantees for a $300 million loan for the startup Crusoe and promising to "backstop" chip rentals when it has no customers

- Google and AMD are adopting strategies to compete with NVIDIA by building their own AI chip ecosystems, focusing on funding cloud partners to use their chips.  
- Google is reportedly planning a $100 million investment in the cloud startup Fluidstack to enhance its AI capabilities, while AMD is guaranteeing a $300 million loan for the startup Crusoe to purchase its AI chips.  
- Both companies are circumventing traditional cloud giants due to their lack of interest in Google's TPU, prompting them to support emerging cloud firms to gain market traction.

Wallstreetcn

Invesco Semiconductors ETF

Roundhill GOOGL WeeklyPay ETF

Invesco PHLX Semiconductor ETF

Hut 8 Mining

Global X Cloud Computing ETF

Direxion Semicon Bull 3X

ISHRS S&P Glb It

SPDR S&P Semicon

Alphabet - C

VanEck Semiconductor ETF

iShares Semiconductor ETF

Direxion Daily AMD Bull 2X Shares

Alphabet

Everyone wants to learn from Nvidia's "chip-for-financing," Google and AMD both want to support "AI cloud"