Meta Enters Three-Year Deal to Utilize Amazon Custom AI Chips

Illustration for: Meta Enters Three-Year Deal to Utilize Amazon Custom AI Chips
AI-generated illustration. Visual interpretation does not represent real individuals or scenes.

THE BARE STORY

Meta Platforms has agreed to a minimum three-year partnership to utilize Amazon Web Services' Graviton processors for its computing and artificial intelligence workloads. The agreement establishes Meta as a top-five customer for the custom central processing units, deploying tens of millions of cores to help reduce the operational expenses of running AI applications across its platforms.

The AWS Graviton chips are built using architecture developed by Arm Holdings, which collects royalties on each deployed unit. Following the announcement on Friday, Amazon shares increased by nearly three percent to over $263. Arm Holdings also experienced significant market movement, with its stock rallying more than 30 percent over the course of the week.

The partnership reflects a broader industry trend of integrating more central processing units alongside traditional graphics processing units to manage AI clusters and improve cost efficiency. According to Amazon Chief Executive Officer Andy Jassy, customers are increasingly seeking alternatives to rely less exclusively on conventional graphics processors. Jassy stated that Amazon’s own chip division is growing at a triple-digit percentage year-over-year, reaching an annual revenue run rate of over $20 billion.

Further details regarding the technology sector's infrastructure investments are expected in the coming week, as major firms including Amazon, Meta, Alphabet, and Microsoft are all scheduled to release their quarterly earnings reports.

Same Facts. Different Perspectives.

Two AI models. Two viewpoints. One factual foundation.

• Accelerating Mega-Corporate Consolidation The Left views the three-year, tens-of-millions-of-cores partnership between Meta and Amazon not as a standard business contract, but as a dangerous concentration of technological power. By cementing an alliance between two industry behemoths, the deal further entrenches monopoly-like conditions within the internet ecosystem. Consumer advocates warn that this level of infrastructure dependency stifles smaller competitors who cannot match the economy of scale these top-five technology giants are building.

• Enriching Elite Capital Holders Prioritizing equitable economic outcomes, this camp points to the immediate 30 percent spike in Arm Holdings and Amazon’s $20 billion chip revenue run rate as evidence of upward wealth extraction. The "cost efficiency" celebrated by executives rarely translates to consumer savings or worker benefits. Instead, these operational expense reductions are explicitly engineered to inflate quarterly earnings reports and maximize returns for a highly concentrated class of shareholders and institutional investors.

• Centralizing Critical AI Infrastructure There is deep skepticism regarding the shift toward proprietary, vertically integrated ecosystems like Amazon’s custom AWS Graviton processors. As the industry pivots away from conventional graphics processors toward closed-loop tech silos controlled by firms like Amazon, Meta, Alphabet, and Microsoft, public reliance on unaccountable megacorporations deepens. This concentration creates a systemic vulnerability where a single, profit-driven corporate entity effectively dictates the foundational architecture of future artificial intelligence development.

How it may affect me

As a U.S. reader:

• In the short term, individuals with investments or retirement accounts holding Amazon or Arm Holdings stock may experience financial gains due to the immediate market rally surrounding this deal.

• Over the long term, the consolidation of artificial intelligence infrastructure between massive technology entities may stifle smaller domestic businesses and startups attempting to compete in the industry.

• Although the partnership significantly reduces operational costs for running major internet platforms, these savings are expected to benefit corporate shareholders and investors rather than lower costs for everyday consumers.

• The foundational architecture of future artificial intelligence development will become increasingly centralized, deepening public reliance on the proprietary hardware systems of a small group of large corporations.

Read the story at

Note: All TheBareNews content is AI-generated. For additional context, reporting, and updates, you are invited to explore the news outlets linked above.