Amazon's AI Chip Strategy Targets Nvidia's Dominance In Data Centers

Amazon.com Inc AMZN is ramping up efforts to challenge Nvidia Corp NVDA in the artificial intelligence (AI) chip market, worth over $100 billion.

From a bustling engineering lab in Austin, Texas, Amazon’s engineers are pushing to market its latest AI chip, Trainium 2, as quickly as possible, Bloomberg reports. Rami Sinno spearheads the Trainium 2 project and leads Amazon’s chip design and testing.

Also Read: Gap’s Growth Drivers: Market Share Gains, Brand Momentum, And Raised Guidance – Analysts Weigh In

Despite Nvidia’s significant lead in AI hardware, Amazon’s ambitions to reduce dependency on external suppliers and deliver competitive AI solutions have gained traction.

Amazon’s latest AI chip, Trainium 2, offers improved performance with four times the computational power and three times the memory of its predecessor, per the report.

Engineers have streamlined its design by reducing internal components and improving heat management. The chip is central to Amazon’s strategy of scaling its AI capabilities and offering a cost-effective alternative to Nvidia hardware.

Prominent AI startups, including Anthropic, are adopting Amazon’s Trainium chips for their workloads. Last week, Amazon increased its funding in Anthropic to $8 billion, solidifying its collaboration on AI advancements and positioning Amazon Web Services (AWS) as the startup’s primary cloud provider.

In October, Amazon signed a five-year agreement with Databricks to deliver cost-efficient AI-building solutions to businesses.

The collaboration focuses on Amazon’s Trainium AI chips, offering a budget-friendly alternative to Nvidia’s widely used GPUs for AI customization. Databricks intends to pass on the savings from Trainium’s use to customers, challenging Nvidia’s dominance in the AI hardware market.

Amazon faces hurdles in software development, an area where Nvidia excels. Its Neuron SDK, still in its early stages, must evolve to match the ease and flexibility Nvidia offers to AI developers.

Meanwhile, Nvidia’s Blackwell chips won’t reach the market until early 2025, impacting significant clients such as AWS. Production challenges at Nvidia have forced AWS to await bulk production of Blackwell chips for its data centers. AWS Chief Matt Garman highlighted the delay during an interview on Bloomberg Television, citing Nvidia’s production adjustments.

In September, DA Davidson’s Gil Luria downgraded Microsoft Corp to Neutral, citing diminishing AI and cloud leadership as key factors. Luria highlighted AWS and Alphabet Inc Google Cloud’s strides in silicon deployment, outpacing Microsoft’s Azure in data center advancements. Despite early AI gains, Luria flagged competitors like AWS and GitLab as catching up to GitHub Copilot, challenging Microsoft’s edge in innovation.

Price Action: AMZN stock closed lower by 0.64% at $197.12 on Friday.

Image via Shutterstock

Also Read: