Tired of ads? Enjoy an ad-free experience by signing up.
👩‍🍳 How we use AI at Tech in Asia, thoughtfully and responsibly.
🧔‍♂️ A friendly human may check it before it goes live. More news here

Amazon to invest $50b in OpenAI, expands cloud deal

Amazon plans to invest a total of US$50 billion in OpenAI, starting with US$15 billion and, if certain conditions are met within the next few months, adding US$35 billion.

OpenAI and AWS said they are expanding their multi-year contract from US$38 billion to US$100 billion over eight years, and OpenAI will secure about 2 gigawatts of Trainium capacity on AWS.

The partnership includes the joint development of a stateful runtime environment based on OpenAI models to be offered through Amazon Bedrock. This tool aims to help developers preserve context while accessing AWS data and compute.

Furthermore, the companies will work to customize OpenAI models specifically for Amazon developers to integrate more deeply with AWS infrastructure.

🔗 Source: Chosun Daily

🧠 Food for thought

Implications, context, and why it matters.

Amazon’s investment marks a new phase of AI non-exclusivity

  • Amazon’s US$50 billion stake sits inside OpenAI’s US$110 billion funding round, with US$30 billion each from Nvidia and SoftBank, at a US$730 billion pre-money valuation 1.
  • A revamped Microsoft OpenAI contract in October 2025 dropped Microsoft’s right of first refusal to serve as OpenAI’s compute provider, and eased limits that had blocked OpenAI from co-building products with outside companies 1.
  • Microsoft stays in place. It keeps an exclusive license to OpenAI’s IP and remains the exclusive cloud for stateless OpenAI API calls 1. Amazon becomes the sole third-party cloud distributor for OpenAI’s Frontier enterprise platform, which pushes a multi-cloud play for top AI firms 1.

The deal speeds the shift toward custom chips and agent platforms

  • OpenAI plans to use about 2 gigawatts of Trainium capacity on AWS. That choice backs custom silicon (custom-designed chips) as an option to Nvidia’s GPUs for at-scale AI workloads, and it may widen hardware mix across the sector 2.
  • Work on a “stateful runtime environment” suggests the next race includes platforms for persistent, context-aware AI agents (software that can take actions on a user’s behalf over time), not only access to AI models 3.
  • Enterprises that build complex agents on Amazon Bedrock could face steep switching costs. That friction can make it hard to move integrated systems to a rival cloud provider 4.

Stay ahead in Asia’s tech landscape

You've reached your 2 free content limit for the month. Sign up for free to read the full story.

🏄 For casual readers / 👶 Free

Basic

US$0

Free forever

Get instant access to this article and more every month

0 premium content

Unlimited news briefs

5

5 articles

Ad-free reading experience

Just US$0 per day

⌛Sign up in 20s. No payment details needed.

📖 For learners / 👍 Starter

Lite

US$4.92/month

Billed annually at US$59/year

Get instant access to this article and more every month

4

4 premium content

Unlimited news briefs & articles

Ad-free reading experience

Just US$0.17 per day

Cancel anytime

Our subscriber community includes professionals from these companies: