Copilot vs ChatGPT privacy: the same OpenAI models under two companies' rules

The Morse teamPublished November 9, 2026Facts checked November 9, 202611 min read

On Copilot vs ChatGPT privacy, as of November 9, 2026, Copilot is ahead on training and ChatGPT is ahead on ads and on knowing when your data is gone. Microsoft's new Copilot app doesn't use your chats to train foundation models, and ChatGPT trains on Free, Plus, and Pro chats until you switch it off. ChatGPT keeps ads away from health, politics, and anyone under 18, gives a 30-day deletion window, and works without an account, while Copilot requires a Microsoft account, publishes no deletion period, and can pick ads from your chats, including teenagers' chats. Copilot also answers with OpenAI's models, so choosing between the two often decides which company keeps the transcript more than which model writes the reply.

We read Microsoft's Copilot pages for individuals (privacy controls, activity history, young people, the Transparency Note, and the Copilot Terms of Use), the Microsoft Privacy Statement (last updated September 2026), and its Microsoft Learn pages on enterprise data protection and OpenAI as a subprocessor. From OpenAI we read its help pages on data controls, chat retention, consumer data, temporary chat, memory, and ads, plus its US privacy policy and enterprise privacy page. All were read on November 9, 2026, for personal accounts unless we say otherwise. Microsoft deserves credit for taking its consumer app off foundation-model training without a switch to hunt for, and OpenAI for putting numbers and topic limits on things Microsoft leaves open.

Copilot vs ChatGPT vs Morse, row by row

As of November 9, 2026. The Copilot column is the app Microsoft updated on August 18, 2026, signed in with a personal Microsoft account. The Morse column comes from our privacy policy.

Is Copilot more private than ChatGPT?

On training, yes. On almost everything else, ChatGPT tells you more. Microsoft's activity history page says "Prompts, responses, and your file contents when using the Microsoft Copilot app aren't used to train foundation models" (Microsoft). OpenAI's data controls page says that "On personal workspaces with ChatGPT Free, Plus, or Pro, data sharing is enabled by default" (OpenAI). The full story of ChatGPT's switch, and its gaps, is in does ChatGPT train on your data.

Two caveats keep Copilot's lead narrow. The Microsoft Privacy Statement, which covers every Microsoft product, still says Microsoft "may use your data to develop, train, and fine-tune our AI models, including large language models (LLMs)" (Microsoft), so "foundation models" is the exact boundary. And Copilot's terms grant Microsoft the right to "copy, distribute, transmit, publicly display, publicly perform, edit, translate, and reformat" your content to operate and improve Copilot, adding "we don't have to pay you, ask your permission, or tell you when we do" (Microsoft).

ChatGPT wins on time limits. A deleted ChatGPT chat is scheduled for permanent deletion within 30 days, unless it was "already de-identified and disassociated from your account" or OpenAI "must retain it longer for security or legal obligations" (OpenAI). Microsoft's page for the new Copilot says only that it "keeps your prior conversations" and points you to the privacy dashboard to delete them (Microsoft). How to clear each one is in is Copilot private and how to delete ChatGPT history.

Does Copilot send your chats to OpenAI?

For personal accounts, no Microsoft page we read says so. For work accounts, Microsoft now uses OpenAI as a subprocessor, switched on by default in July 2026. Copilot's Transparency Note says it uses "language and multi-modal models from Microsoft, OpenAI, and other model developers" (Microsoft), and its terms say that "When you use Copilot, you are subject to the Microsoft Privacy Statement" (Microsoft). So OpenAI's models can answer you in either app, while a different company keeps the transcript and sets the ad, review, and retention rules. None of the consumer Copilot pages we read says OpenAI receives or keeps your Copilot chats.

The work version changed this summer. A June 2026 comparison that ranks for this query says Microsoft 365 Copilot is "built on OpenAI's models, delivered through Azure rather than the public ChatGPT product." Microsoft's own page now says it "provides OpenAI operated models through OpenAI as a subprocessor," that OpenAI joined its subprocessor list on June 23, 2026, and that "As of July 24, 2026, OpenAI operated models are enabled for all users for eligible commercial customers" unless an admin selects No users (Microsoft Learn). Microsoft says its Data Protection Addendum and enterprise data protection still apply, and that "OpenAI offers Zero Data Retention for the Responses API used by Microsoft," subject to limits in OpenAI's own documentation. Admins find the switch in the Microsoft 365 admin center under Copilot > Settings > View all > AI providers operating as Microsoft subprocessors.

Which one uses your chats for ads?

Both, on their free tiers, under different rules. Microsoft says "If you use Microsoft Copilot and don't have a Microsoft 365 subscription, you might see advertising in Copilot," with generic ads chosen from "the most recent content of your current conversation" (Microsoft). OpenAI says ads "may appear for users on the Free and Go plans," and Plus, Pro, Business, Enterprise, and Edu accounts "will not have ads" (OpenAI).

The rules differ in three places:

• Teenagers. OpenAI shows no ads to accounts it identifies as under 18. Microsoft's page for young people says they "may see ads while using Copilot," chosen from the chat rather than "other information about you" (Microsoft).
• Sensitive topics. OpenAI says ads "are not eligible to appear near sensitive topics, including personal health, mental health, or politics." Microsoft's Copilot controls page sets no topic rule, and its Privacy Statement says only that Microsoft Advertising doesn't use "sensitive data" under US state laws for personalized ads (Microsoft).
• Reach beyond the chat. Microsoft's personalized Copilot ads can use "other Microsoft data you've chosen to share," and its Privacy Statement lists "your Copilot conversations (including files you share)" among what ads may relate to. OpenAI says the signals behind ChatGPT ads are "information that stays within ChatGPT," though its privacy policy says it "could receive information about purchases you make from these advertisers" to measure ads (OpenAI).

To turn the personal part off, use Settings > Personalization > Allow ads personalization in Copilot, or Settings > Ad Controls in ChatGPT. Neither removes ads. Our guide to turning off ChatGPT ads covers OpenAI's Ads-Free option for the Free plan, which trades ads for lower limits.

Who can read your Copilot or ChatGPT chats?

People at both companies, and OpenAI spells out when. Copilot's terms say it "may include both automated and manual (human) processing of data" (Microsoft), and Microsoft's page for young people lists safety, legal, product improvement, and troubleshooting as reasons. No new-app page we read gives an opt-out or says how long a reviewed chat is kept.

OpenAI names four reasons staff and "trusted service providers" may access your content: abuse or security investigations, support you ask for, legal matters, and model improvement "unless you have opted out" (OpenAI). It says it monitors and logs "all access to user content." Turning training off removes one of the four. Both privacy policies also let the company disclose data when the law requires it, which we cover in can AI chats be subpoenaed.

What does each account tie your chats to?

Copilot ties them to the Microsoft account you also use with Bing, Edge, and MSN, and ChatGPT to an OpenAI account you can skip. Copilot's One shared experience toggle lets your chats shape Bing and MSN. Microsoft's example is that a chat about "future vacation ideas" brings "content tailored to your upcoming plans" (Microsoft). In the US, Copilot can also import your Edge "cookies, history, payment info, passwords, and autofill data" if you turn that on under Settings > Web browsing.

Memory works much the same in each, with the same gap. Turning off Saved memories in Copilot "doesn't automatically delete already saved memories," and in ChatGPT, "Deleting a chat alone does not necessarily delete a separate saved memory created from that chat" (OpenAI). ChatGPT's Memory can also draw on connected apps such as Gmail. Clearing it fully is in how to turn off ChatGPT memory.

Copilot or ChatGPT for sensitive data at work?

Either, on a work account, with your employer able to see it. Both companies keep business chats out of training by default and give admins the history.

Microsoft's work pages note one more gap: "Models provided by Anthropic as a subprocessor are currently excluded from the EU Data Boundary" (Microsoft Learn). A personal account on a work laptop is a different problem, covered in AI on a work laptop, and OpenAI's business plan has its own page, is ChatGPT Business private. This isn't legal advice.

How Morse lets you use GPT without a Microsoft or OpenAI account behind it

Every row in the first table describes a copy of your chat kept under an account: a Microsoft account that also runs your search, browser, and ad settings, or an OpenAI account. Choosing between the two decides which company's rules apply to that copy, and both may answer with OpenAI's models. Morse is built so the copy isn't on anyone's server under your name, ours included.

• Anonymous, to ask GPT without Microsoft or OpenAI knowing it's you. You pick GPT, Claude, or Gemini, and our proxy strips your account, email, and IP address before sending the message under Morse's own account. The provider reads it without knowing it's you, and may keep it for a limited time to check for abuse, under its own policy. No Microsoft account means no shared experience with Bing and no ad profile. Morse removes who you are, not what you type, so leave names out.
• E2EE, for the sensitive data you wouldn't give either company. Your browser checks the secure enclave's hardware attestation, then encrypts the message so only that enclave can read it, and Morse's servers relay ciphertext they can't read. The enclaves are run by third-party providers, not by Morse, and the guarantee rests on the enclave hardware and attestation working as designed. E2EE runs open-weight models like GLM, Kimi, and DeepSeek, without web search or image generation, and comes with Pro at $19 a month.
• Private, the default, for daily questions with nothing kept at the host. Open-weight models and xAI's Grok run under Morse's zero data retention agreements with its inference, search, and other providers. Where a provider confirms zero retention on each reply, Morse checks it and refuses a reply that doesn't.

Whatever the mode, there's no training switch, because Morse doesn't train AI on your content, and no ad settings, because it shows no ads and sells no data. It runs no advertising trackers or third-party analytics. Your history is stored in your browser, encrypted, locked when you sign out, and opened with two keys: one that never leaves your device and one our server gives you at sign-in. Because we don't store chats, we usually have very little to give authorities. Deleting means deleted, with no 30-day copy on our side, and clearing your browser removes your chats for good, so keep a copy of anything you need. The privacy modes page shows who sees what in each mode.

Choose Copilot, ChatGPT, or Morse

• Choose Copilot if you already live in Windows, Edge, and Microsoft 365, want a consumer app that keeps your chats out of foundation-model training without a setting, and pay for Microsoft 365, since Microsoft's ad warning names only people without a subscription.
• Choose ChatGPT if you want stated deletion windows, ads kept away from health and politics, no ads for teenagers, and the option to chat signed out, and you'll switch off training yourself.
• Choose Morse if you want GPT without a Microsoft or OpenAI account behind it, open-weight models on hosts that agreed to keep nothing, E2EE your own browser verifies, and history kept encrypted on your device, with no training or ads on any plan.

Send a first message from the Morse home page with no account, or start free with 15 messages a day and no card.

Morse publishes this comparison. We've tried to keep it fair; if something's wrong or out of date, email hello@trymorse.ai.

Copilot is a trademark of Microsoft Corporation. Morse is not affiliated with or endorsed by Microsoft Corporation.

ChatGPT is a trademark of OpenAI. Morse is not affiliated with or endorsed by OpenAI.