On 9 November 2026, OpenAI put ChatGPT Enterprise on sale through AWS Marketplace. If your company already buys software through Amazon Web Services, you can now buy the workplace version of ChatGPT the same way — on your existing AWS bill — and count eligible spending toward the cloud-spending commitments you have already signed with Amazon. There is one catch to know before you start: there is no public price, and the listing’s own fine print says buying through AWS does not guarantee your data stays on AWS infrastructure. This guide walks through exactly how the purchase works, what it costs, and the five questions to answer before you sign.

Quick answer: ChatGPT Enterprise is now on AWS Marketplace: buy it on your AWS bill and count eligible spend toward EDP or PPA commitments. Here's how the purchase works.

What changed on 9 November 2026

OpenAI announced that ChatGPT Enterprise is available through AWS Marketplace, giving organisations a way to buy and manage the suite through their existing AWS purchasing and billing relationships and apply eligible spending toward AWS commitments. That is the whole announcement, in substance: a new way to buy, not a new product. OpenAI says customers buying through AWS get the same models, features and functionality as customers who buy directly from OpenAI, with new features released at the same time. The listing names OpenAI as the seller and AWS as the party that handles purchasing and billing.

This is the second half of a two-part arrangement with Amazon. On 28 April 2026, OpenAI and AWS announced limited previews of OpenAI models on Amazon Bedrock, Codex on AWS and Amazon Bedrock Managed Agents powered by OpenAI — the half for developers building with OpenAI models inside AWS. The 9 November listing is the other half: packaged workplace AI for the rest of the business, bought through the same cloud storefront. As OpenAI put it, the two “complement” each other: developers build in Bedrock, teams across the business use ChatGPT Enterprise in their daily work. OpenAI opened its announcement with scale: more than 1.2 billion people use ChatGPT each week, according to the company — its own reach claim for ChatGPT as a whole, not a count of Enterprise seats.

What you actually get: four products in one listing

The AWS Marketplace listing bundles four OpenAI products — Chat, ChatGPT Work, Codex and dots — into one enterprise offering. They cover different kinds of work, and it helps to know which one your team will actually use before you pay for all four.

Product What it does Who it is for
Chat The conversational product most people know: explore ideas, ask questions, write and work through problems in conversation. Every employee
ChatGPT Work Takes on multi-step research, analysis and projects, using connected apps and files to produce documents, presentations, spreadsheets and reports; users review progress, steer the work and approve important actions. Knowledge workers running longer projects
Codex OpenAI’s coding agent: understands codebases, builds features, fixes bugs, runs tests and reviews changes. Developers
dots Always-on agents that track updates, keep documents and analyses current, and bring forward results and decisions that need a person’s attention between conversations. Teams with ongoing responsibilities

Two details are worth noting. ChatGPT Work is the newest of the four: OpenAI introduced it on 9 July 2026, when it began folding its standalone Atlas browser into the new agent. And the listing gives no limits on how many dots an organisation can run, what they cost, or how their activity is metered — those are questions for the private offer, not the product page. Eligible customers can also access ChatGPT for Healthcare and ChatGPT for Financial Services through the same Marketplace channel, with eligibility confirmed through an OpenAI or AWS account team; neither the announcement nor the listing publishes the qualification criteria.

How to buy ChatGPT Enterprise through AWS Marketplace

There is no self-service checkout. Every deal is a private, negotiated quote, so treat the steps below as a procurement process, not a shopping trip.

  1. Check your AWS commitment first. The main reason to buy through AWS is that eligible spending can count toward your AWS Enterprise Discount Program (EDP) or Private Pricing Addendum (PPA) commitment — the multi-year minimum-spend deals large organisations sign with Amazon in exchange for discounts. Find out how much headroom your commitment has, and what share of it Marketplace purchases can consume: the listing says eligible consumption counts toward commitments “subject to” your agreement with AWS, which means the exact treatment is set contract by contract. If you have no EDP or PPA, this route buys you nothing extra.
  2. Find the listing in AWS Marketplace. Search for ChatGPT Enterprise in the AWS Marketplace console. The listing is filed under Software Development, Research and Generative AI, and is delivered as software as a service — the software runs as a hosted service managed by OpenAI, not inside your own AWS environment, despite the page’s “Deployed on AWS” label.
  3. Request a private offer. Under “Pricing”, the listing shows only “Custom pricing options” and states that pricing depends on your requirements and eligibility. No per-seat price, minimum seat count, contract length or usage rate is published. Request the private offer and expect a custom quote. At the time of the announcement the listing showed no customer ratings or reviews.
  4. Confirm the commitment drawdown in writing. Before signing, get your AWS account team to confirm in writing that your ChatGPT Enterprise spend will count toward your EDP or PPA, and how much of it counts. This is the financial heart of the deal — the whole point of buying through AWS — and it is governed by your agreement with AWS, not by the listing page.
  5. Confirm eligibility for vertical versions if you need them. If you want ChatGPT for Healthcare or ChatGPT for Financial Services, confirm eligibility with an OpenAI or AWS account team before you sign. Note that the ChatGPT Enterprise listing itself makes no statement about HIPAA, the US health privacy law; OpenAI points healthcare buyers to the dedicated Healthcare version instead.
  6. Set up admin controls before rollout. The listing says business data is encrypted at rest and in transit and is not used to train OpenAI’s models by default; administrators can manage user access, permissions and connected tools, while users control what agents can access and which actions require approval. Turn those controls on during setup, not after the first incident — connected tools for enterprise workspaces are off by default and require an administrator to enable them.

What it costs — and what to ask before you sign

Nobody outside the negotiation knows the price, because OpenAI has not published one. What is public is the mechanism: OpenAI is the seller, AWS handles the bill, and buyers pay recurring monthly usage fees through their AWS invoice. How credit consumption inside ChatGPT Enterprise translates into the charges on that invoice is not explained in the announcement materials — OpenAI introduced usage analytics and updated spend controls for enterprises in June 2026, with default, group and individual limits in one admin view, but the AWS materials do not say how those credits map to Marketplace billing. Three questions belong in every quote request: what share of our EDP or PPA commitment does this spend draw down; what is the minimum commitment, and for how long; and what is included in the quote versus metered separately — dots, in particular, have no published limits or metering in the listing, so ask before you deploy them widely.

Neither company disclosed what share of each sale AWS retains. For scale, Anthropic’s amended IPO filing put sales through the cloud marketplaces of Amazon, Google and Microsoft at about 11% of its revenue in 2023, 32% in 2024, 47% in 2025 and 44% in the nine months to 30 September 2026. OpenAI has not disclosed a comparable figure, and the announcement does not say whether ChatGPT Enterprise is also listed on the marketplaces of Microsoft or Google.

The data catch: what the listing does — and does not — promise

Buying through AWS Marketplace does not guarantee that all of your data stays on AWS infrastructure or within your AWS environment. The listing says so explicitly, and this is the single most important sentence in the fine print for any regulated or privacy-conscious buyer.

What the listing promises What it does not promise
Business data encrypted at rest and in transit That all customer data remains on AWS infrastructure
Business data not used to train OpenAI models by default That data stays within your own AWS environment
OpenAI is the data processor; AWS is a subprocessor Data or inference residency in your preferred region
Admin control over user access, permissions and connected tools That residency controls cover all processing or systems data

The two residency controls are separate, and the distinction matters. Data residency controls where supported customer content is stored at rest. Inference residency separately controls where model inference on that content runs. Both options apply only when enabled, are not available in all regions and do not cover all processing or systems data. Zeniteq’s reporting adds a sharper detail: for AWS Marketplace workspaces, inference residency can be selected only in the United States — a claim we could not verify against OpenAI’s own documentation at the time of writing, so treat it as reported, not confirmed, and ask your account team to confirm it in writing if your data must be processed outside the US. For a fuller picture of how to think about data controls before switching any AI tool on, see our guide to opting out of AI training with robots.txt.

Buy through AWS or buy direct from OpenAI?

If you have an AWS commitment with headroom, buy through AWS. If you do not, buy direct. The software is the same either way, so the trade-offs are about procurement, not product.

Your situation Buy through AWS Marketplace Buy direct from OpenAI
You have an EDP/PPA with unspent headroom Yes — the drawdown is the whole point. Eligible spend counts toward a commitment you have already promised to meet. You pay the same money but burn none of your AWS commitment.
You have no AWS commitment No financial benefit; you add a middleman to the purchase. Simpler. One contract, one vendor, one bill.
You need data processed outside the US Check first: inference residency for Marketplace workspaces is reported to be US-only. Ask OpenAI directly about residency options for direct workspaces.
You run a small team with no AWS account team The private-offer process is built for large buyers; expect friction. Simpler. Direct sales is the route built for you.
You already buy most software through AWS Yes. As AWS’s Matt Yanchyshyn put it, customers can “buy it through their AWS account, govern it alongside the rest of their software spend, and move from evaluation to rollout faster.” You add a second procurement channel to govern.

One more comparison is worth making. Anthropic’s Claude Enterprise is already sold the same way on the same marketplace — the listing’s “Similar products” panel shows it alongside Perplexity Enterprise Pro — and its description says that for AWS customers, spend draws down directly from an EDP or PPA commitment. OpenAI’s offer arrives with the same commitment mechanism its main rival already uses. AWS now sells the workplace products of both leading model developers through the same storefront, and it collects a fee on every sale regardless of which model wins the seat. If you are weighing the two, our Claude Chat and Cowork enterprise guide covers the rival’s setup, permissions and the 16 November switch in detail.

ChatGPT Enterprise or Amazon Bedrock?

They are different purchases for different people, and OpenAI says they complement each other. Bedrock is where developers build AI applications and agents with OpenAI models — GPT-6 Astra, GPT-6 Sol and others are available there, alongside Codex. ChatGPT Enterprise is the packaged workplace product that staff use in their daily work. If your question is “how do we let our developers build with these models”, the answer is Bedrock. If your question is “how do we give 500 employees a governed ChatGPT”, the answer is this listing. Many large organisations will end up with both, governed by different teams, which is exactly why the data-residency and admin-control questions above deserve a written answer before either one is signed.

Our take: the commitment is the product

The honest way to read this announcement is that OpenAI is selling a procurement shortcut, not a new capability — and the shortcut is only valuable if you have a cloud commitment to burn. The software is identical to the direct version, the price is unpublished, and the data-residency fine print is weaker than the “Deployed on AWS” label suggests. None of that makes the listing a bad deal; it makes it a deal whose value lives entirely in your AWS contract. Before you request a private offer, answer these five questions:

  1. Do we have an EDP or PPA with headroom, and what share of it can Marketplace purchases consume?
  2. Is US-only inference residency acceptable for the data our teams will put into ChatGPT?
  3. Do we need the Healthcare or Financial Services version, and is our eligibility confirmed in writing?
  4. Have we priced the direct route from OpenAI for comparison — same product, one fewer intermediary?
  5. Who owns the AWS bill internally, and will they govern this purchase alongside the rest of our software spend, as AWS’s pitch assumes?

If the answer to the first question is no, stop here and buy direct. If it is yes, the Marketplace route is the cheaper way to buy the same product — just get the drawdown and the residency answers in writing first. Desk-researched, not hands-on tested: we have no access to a private offer or an AWS Marketplace listing account, so the steps above follow the published listing and announcement rather than a completed purchase, and we have said so wherever a claim rests on a vendor document rather than a test.

Frequently Asked Questions

Can I buy ChatGPT Enterprise through AWS Marketplace right now?

Yes, if you are an organisation buying through AWS. OpenAI announced the listing on 9 November 2026. There is no self-service checkout: you request a private offer from the AWS Marketplace listing and receive a custom quote based on your requirements and eligibility.

How much does ChatGPT Enterprise cost on AWS Marketplace?

There is no public price. The listing shows only “Custom pricing options”, with no per-seat price, minimum seat count, contract length or usage rate published. You must request a private offer for a quote, and you should ask how the spend counts toward your AWS EDP or PPA commitment.

Does buying through AWS Marketplace count toward my AWS cloud commitment?

Eligible spending can count toward your AWS Enterprise Discount Program or Private Pricing Addendum commitment, subject to your agreement with AWS. The exact treatment is set contract by contract, so confirm it in writing with your AWS account team before signing.

Is my data kept on AWS if I buy ChatGPT Enterprise through AWS Marketplace?

Not necessarily. The listing states that purchasing through AWS Marketplace does not guarantee all customer data remains on AWS infrastructure or within your AWS environment. Data residency (where content is stored at rest) and inference residency (where inference runs) are separate controls, region-limited, and do not cover all processing. Business data is encrypted at rest and in transit and is not used to train OpenAI models by default.

Is ChatGPT Enterprise on AWS Marketplace the same product as buying direct from OpenAI?

Yes, according to OpenAI: the same models, features and functionality, with new features released at the same time. The bundle is Chat, ChatGPT Work, Codex and dots, and eligible customers can also access ChatGPT for Healthcare and ChatGPT for Financial Services. What changes is how you buy and pay — OpenAI sells, AWS bills — not what the software does. For the security-model side of OpenAI’s enterprise push, see our GPT-6 Cyber explainer.

Sources and Methodology

Sources: OpenAI, “ChatGPT Enterprise is now available through AWS Marketplace” (announcement, 9 November 2026, openai.com); Unite.AI, “OpenAI Starts Selling ChatGPT Enterprise Through AWS Marketplace” (9–10 November 2026, unite.ai); PPC Land, “AWS customers can now buy ChatGPT Enterprise against cloud commitments” (9–10 November 2026, ppc.land); Zeniteq, “ChatGPT Enterprise Joins AWS Marketplace” (10 November 2026, zeniteq.com); NGlobeNews, “OpenAI lists ChatGPT Enterprise on AWS Marketplace” (9 November 2026, nglobenews.com).

Method: desk research on 11 November 2026. Every factual claim was checked against OpenAI’s own announcement as reported by Unite.AI and PPC Land, both read in full this run, with Zeniteq and NGlobeNews as independent cross-checks. No claim rests on a single source where a second source was available; vendor claims (pricing mechanism, feature parity, user counts) are attributed to the vendor. The US-only inference-residency detail is reported by Zeniteq only and is labelled as such. No hands-on purchase was made — the author has no access to a private offer or an AWS Marketplace listing account — and the article says so plainly. Internal links point to openaimaster.ai’s Claude enterprise guide, GPT-6 Cyber explainer and AI-training opt-out guide.

Leave a Reply

Your email address will not be published. Required fields are marked *