| Ticker | Status | Jurisdiction | Filing Date | CP Start | CP End | CP Loss | Deadline |
|---|
| Ticker | Case Name | Status | CP Start | CP End | Deadline | Settlement Amt |
|---|
| Ticker | Name | Date | Analyst Firm | Up/Down | Target ($) | Rating Change | Rating Current |
|---|
OpenAI is accelerating its push to dominate the artificial intelligence landscape with a staggering $500 billion chip initiative involving Broadcom Inc. (NASDAQ:AVGO).
On Monday, the company behind ChatGPT struck a multibillion-dollar agreement with Broadcom to co-design custom chips capable of powering its AI models, marking one of the most extensive semiconductor procurement campaigns in history.
According to the Financial Times, the deal could cost OpenAI between $350 billion and $500 billion, bringing its total chip capacity plans to over 26 gigawatts—comparable to the energy output of 26 nuclear reactors.
Chief Executive Sam Altman said the company has worked with Broadcom for 18 months to develop silicon tailored for AI inference, the computational process of responding to user queries.
“The biggest joint industrial project in human history” is how Altman described the AI infrastructure push.
Broadcom CEO Hock Tan compared the AI revolution to the advent of the internet or railroads, calling it a "critical utility over time for 8 billion people globally."
Shares of Broadcom Inc. rallied as much as 9% during premarket trading in New York, on track to fully recoup losses from Friday.
See Also: China’s Silence Is Scarier Than Tariffs—5 Ways It Could Hit Trump Hard
The Broadcom pact follows two other major deals OpenAI recently signed in its AI hardware arms race. In September, OpenAI committed to a 10-gigawatt order from Nvidia.
Last month, it added another 6 gigawatts with AMD, bringing the total across all three vendors to 26 gigawatts.
Nvidia's involvement includes a $100 billion equity investment into OpenAI, a move some analysts view as financially circular. James Schneider, an analyst at Goldman Sachs, noted that the structure resembles “circular revenue,” with Nvidia funding OpenAI only to recognize revenue when OpenAI purchases its GPUs back.
INvidia expects $13 billion in revenue from OpenAI in 2026. The company will reinvest $10 billion of its gross profit. Schneider said, "When equity investment comes from a supplier, additional scrutiny is warranted."
Meanwhile, AMD's agreement with OpenAI could be even more transformational. The deal includes up to 6 gigawatts of AMD Instinct GPU deployments, starting with 1 gigawatt of MI450 GPUs in late 2026. AMD will also issue OpenAI up to 160 million shares through performance-linked warrants—potentially worth $75 billion—if deployment milestones are met, including a $600 share price.
Goldman Sachs estimates the AMD deal represents a $135 billion revenue opportunity based on expected GPU pricing. AMD CFO Jean Hu said the partnership has the potential to generate “tens of billions of dollars in revenue” for the chipmaker.
Despite its relatively modest revenue base, OpenAI is committing to over $1.5 trillion in semiconductor and infrastructure spending over the next decade. That includes $300 billion in data center build-outs with Oracle Corp. (NYSE:ORCL).
Each gigawatt of AI capacity requires about $50 billion in total investment—$35 billion for chips and $15 billion for infrastructure. Broadcom's chips are expected to be cheaper than Nvidia's, but even so, the economics are eye-watering.
The company's strategy signals a shift toward vertical integration, involving the design of its own chips, securing dedicated compute power, and minimizing reliance on third-party vendors.
Still, the sheer size of the investment raises critical questions.
Now Read:
© Jeff Lange / USA TODAY NETWORK / USA TODAY NETWORK via Imagn Images
