Claude vs Gemini privacy: what Anthropic's and Google's own pages say

The Morse teamPublished November 7, 2026Facts checked November 7, 20268 min read

In a Claude vs Gemini privacy comparison, Claude wins on training and on human review, and Gemini wins on how long it keeps a throwaway chat and on letting you stay unnamed. As of November 7, 2026, Anthropic keeps its training switch separate from your chat history and keeps staff out of your chats by default, while Google trains on your chats by default if you're 18 or over and sends a sample to human reviewers, including its service providers. But Claude demands a phone number from every new user, and Gemini keeps a temporary chat for 72 hours where Claude keeps an incognito chat for 30 days.

We put the same questions to both companies' privacy policies and help pages, all read on November 7, 2026. Both deserve credit for putting a number on nearly every retention period instead of hiding behind "as long as necessary." Still, neither app is private from the company running it.

Claude vs Gemini privacy, question by question

Each row asks one question and gives each company's own answer, then who wins it. Morse is in the last column for comparison.

Claude takes five rows, Gemini three, and one is a tie.

Does Claude or Gemini train on your chats?

Both can, and Claude makes it easier to say no. Anthropic asks new users to choose at sign-up, and its switch, Help Improve our AI models, sits on its own under Settings, then Privacy (Anthropic). Gemini trains by default for adults through Keep Activity, and turning that off also turns off your saved history and most connected apps (Google).

Verdict: Claude. Its training switch is separate from your history. On either app, rating a reply hands some of your chat back for training, so skip the thumbs. The per-setting detail is in is Claude private and is Gemini private.

Who reads your chats, Anthropic's staff or Google's reviewers?

Fewer people read Claude chats. Anthropic says its employees can't open your conversations by default, with exceptions for feedback you send and for chats flagged under its Usage Policy (Anthropic). Google sends "a subset of chats" to human reviewers, including its service providers, with Keep Activity on or off (Google).

Verdict: Claude. A Claude chat needs a flag or your feedback before anyone looks, while any Gemini chat can land in Google's sample. Where either company cuts the link to your account before review, the names you typed stay in the text. We cover reviewers app by app in who reviews AI chats.

Which keeps your chats longer, Claude or Gemini?

Claude keeps them longer in the two cases most people hit. With training on, Anthropic may keep your chats for up to 5 years in its training pipelines, while Gemini's history auto-deletes after 18 months by default (Anthropic; Google). For a one-off question, Claude's incognito chat stays on Anthropic's servers for 30 days, and Gemini's temporary chat for 72 hours (Anthropic; Google).

Verdict: Gemini, for throwaway chats and for default history. Its reviewed copies are the exception, at up to three years. To clear what Gemini already has, see how to delete Gemini history.

Is Claude or Gemini better at keeping your name off your chats?

Gemini, because you can use it signed out, and Google says signed-out chats aren't tied to a Google Account (Google). Anthropic requires a phone number from every new user, with no way to skip it (Anthropic); we compare that rule with other apps' in which AI chatbots need a phone number. Signed in, the picture flips. Gemini sits inside the Google Account that holds your mail and photos, and Google says what Gemini draws from connected apps can train its models "for everyone" (Google). Our page on whether Google uses Gmail for ads shows how to disconnect them.

Verdict: Gemini signed out; Claude signed in. Claude ties your chats to one phone number. A signed-in Gemini ties them to the rest of your Google life.

What happens when the police ask Anthropic or Google?

It's a tie on policy. Both companies allow disclosure to meet an "enforceable governmental request," and both say they tell you about a request unless the law forbids it. Anthropic lists other rare exceptions, such as a child at risk or an emergency, and Google says it emails you before it discloses (Anthropic policy and notice page; Google). Google's count of requests is far larger and covers every Google product, with Gemini not counted separately, as our AI transparency reports comparison shows, and both hold chats that a court can order them to produce.

Gemini or Claude for sensitive information?

Neither, if no company should read it. If you're choosing one anyway, Claude with Help Improve our AI models off and an incognito chat keeps reviewers away unless the chat gets flagged. Gemini with a temporary chat clears it sooner but leaves it open to sampled review. Don't rate replies in either, and don't connect Gmail or Photos to Gemini for anything private. Google's hub itself asks you not to enter confidential information you wouldn't want a reviewer to see (Google).

How Morse lets you use Claude and Gemini without either company knowing it's you

Both companies answer every question above from the same starting point: they hold your chat on their servers, under an account opened with your phone number or your Google identity. Their settings only decide how long that copy lasts and who may read it. Morse is a web app with over 30 models, open and closed, built so your chat history stays in your browser, encrypted, and never on our servers. Each chat has a privacy mode:

• Anonymous, for Claude and Gemini without Anthropic or Google knowing it's you. Our proxy removes your account, email, and IP address and sends the message under Morse's own account. You need no phone number for Anthropic and no Google Account. The provider sees the message without knowing it's you, and may keep it for a limited time to check for abuse, under its own policy. Since it still sees your words, leave out names and details that point to you. The Free plan includes Claude Haiku 5.5 within 15 messages a day, and Pro adds unlimited Claude Sonnet 5.5.
• Private, the default, when you'd rather no host kept the chat at all. Open-weight models like DeepSeek, Kimi, and GLM, plus xAI's Grok, run under zero data retention: Morse has 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.
• E2EE, for the sensitive questions neither company should see. Before anything leaves your device, your browser verifies the enclave's hardware attestation and seals the message for that enclave alone. What passes through our servers is ciphertext we can't read, and the reply opens only in your browser. 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, without web search or image generation, and comes with Pro at $19 a month. Read how end-to-end encrypted AI chat works.

In every mode, Morse doesn't train AI on your chats, sell your data, or show ads. Morse's servers keep no 30-day or 72-hour copy and send nothing to reviewers, because they don't keep messages. In Anonymous, the provider may still keep a message for a limited time to check for abuse. Your history is locked when you sign out and takes two keys to open, 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 a chat or clearing your browser removes it for good, and we can't recover it, so save a copy of anything you need. The privacy modes page shows who can see what in each mode.

Choose Claude, Gemini, or Morse

• Choose Claude if you want to switch training off and keep your history, and you want people kept out of your chats unless something gets flagged, and you're fine giving Anthropic a phone number. As of November 7, 2026, Claude Pro is $20 a month, Max $100, and Ultra $200 (Anthropic).
• Choose Gemini if you want throwaway chats gone in 72 hours, the option to use it signed out, and Gemini working inside Gmail and Photos, and you accept that a sample of chats goes to reviewers for up to three years. Google AI Plus is $4.99 a month, Pro $19.99, and Ultra from $99.99 (Google).
• Choose Morse if you want Claude and Gemini without either company knowing it's you, open-weight models on hosts that keep nothing, E2EE your own browser verifies, and history kept encrypted on your device, while Morse itself never trains on your chats or shows ads.

Try a chat from the Morse home page without an account, or start free with no card needed.

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

Claude is a trademark of Anthropic. Morse is not affiliated with or endorsed by Anthropic.

Gemini is a trademark of Google LLC. Morse is not affiliated with or endorsed by Google LLC.