I'm LongbridgeAI, I can summarize articles.Despite CEO Dario Amodei's call to slow AI development, Anthropic has committed $517 billion to compute power over the past 11 months. This massive spending benefits key partners: Nvidia and AMD supply GPUs for training and inference; Alphabet and Broadcom provide custom TPUs; and Amazon receives over $100 billion in AWS cloud commitments. These investments highlight significant revenue opportunities for major tech firms driving AI infrastructure expansion.
Anthropic CEO Dario Amodei recently published an essay calling for frontier AI labs to slow their development. The worry is that AI could eventually become dangerous without proper safeguards. Movies like The Terminator certainly show what a worst-case scenario could look like.
However, the company's actions tell a much different story. According to technology publication The Information, over the past 11 months through the end of August, Anthropic agreed to spend a whopping $517 billion on computing power. That's well above the $180 billion in commitments it had previously disclosed.
Let's look at five AI stocks set to benefit from this spending.
Image source: Getty Images.
Nvidia
Nvidia's (NVDA +0.57%) graphics processing units (GPUs) remain the primary chips powering AI workloads, so the more money a company like Anthropic spends on AI infrastructure, the more Nvidia benefits. Nvidia remains the dominant player in AI model training, as most foundational AI code was written on its CUDA software platform and optimized for its GPUs. Meanwhile, with its "acquisition" of Groq, it has established itself as an important player in inference.
NASDAQ: NVDA
Key Data Points
Nvidia CEO Jensen Huang recently forecast that AI infrastructure spending would climb to between $3 trillion and $4 trillion by 2030, so the chip giant clearly sees no slowdown in spending, despite Anthropic's recent call for frontier labs to slow the pace of development. Meanwhile, it has been reported that Nvidia could invest up to $10 billion into the frontier lab as an anchor investor in its IPO.
AMD
Nvidia's GPUs aren't the only chips Anthropic is using in its AI infrastructure build-out. Advanced Micro Devices (AMD +2.19%) agreed to invest up to $5 billion in the company, while Anthropic will buy AMD's Helios rack-scale system to run inference.
NASDAQ: AMD
Key Data Points
Helios includes both AMD's new Instinct GPUs and its central processing units (CPUs). The two companies will also work together on Anthropic's Claude model to optimize workloads for its GPUs and improve ROCm software development.
As part of the deal, Anthropic will deploy up to 2 gigawatts of AMD's GPUs, with the rollout beginning in 2027. This follows two other large GPU deals for inference that AMD has made with OpenAI and Meta Platforms.
Alphabet and Broadcom
In addition to GPUs, Anthropic is also using custom AI chips in its data centers. The company is set to become Broadcom's (AVGO -1.58%) largest customer, as it aggressively deploys the Tensor Processing Units (TPUs) that Alphabet (GOOGL -1.26%) (GOOG -1.24%) co-designed with Broadcom's help. This is set to be a big windfall for both Alphabet and Broadcom.
The bulk of the revenue will accrue to Broadcom, which will deliver the physical chips and provide associated networking services. However, Alphabet will generate a high-gross-margin revenue stream that should largely flow through to the bottom line.
Broadcom has said that Anthropic will deploy 1 gigawatt of Alphabet's Ironwood TPUs this year and 5 gigawatts of its TPU v8i in 2027. It then sees another 10 gigawatts of Anthropic TPU deployments in 2028.
Amazon
Another big Anthropic compute provider is Amazon (AMZN -2.02%). The e-commerce and cloud computing company was an early backer of Anthropic and owns around a 20% stake in the frontier lab. Last year, it brought online a massive data center powered by nearly half a million of its Tranium chips exclusively for Anthropic, but that is just the tip of the iceberg.
NASDAQ: AMZN
Key Data Points
In conjunction with its investments over the years, Amazon has received more than $100 billion in Amazon Web Services (AWS) cloud commitments from Anthropic over the next decade. The large language model (LLM) maker will use both Amazon's Trainium AI accelerators and its Graviton CPUs in its deployments.