ChatGPT alternatives ranked by privacy, ads, and what they still refuse

The Morse teamPublished October 25, 2026Facts checked October 20, 20269 min read

If you want a ChatGPT alternative private enough that no company trains on your chats, keeps them on its servers, or shows you ads, none of the big-name assistants clears all three, and each still refuses some things ChatGPT refuses. As of October 20, 2026, ranked on privacy, ads, and refusals, the order is Morse, Proton's Lumo, Duck.ai, Claude, Mistral's Le Chat (now called Vibe), Gemini, and Perplexity.

We read each assistant's own privacy page, usage rules, and pricing page on October 20, 2026, and OpenAI's help and pricing pages on October 19, 2026. The ranking weighs three things: what the company keeps and trains on, whether it shows or measures ads, and what its written rules tell the model to refuse. Every big name on this list does at least one of these better than ChatGPT, and Claude and Lumo do several.

The short list

As of October 20, 2026.

Why people look for a ChatGPT alternative

Three reasons come up, and OpenAI's own pages confirm each one.

Ads. OpenAI's ads help page says "Ads may appear for users on the Free and Go plans," and the pricing page says of the $8 Go plan, "This plan may include ads" (OpenAI, OpenAI), as of October 19, 2026. Getting rid of them means Plus at $20 a month, a Temporary Chat, or the Ads-Free option with lower limits. We cover each route in how to turn off ChatGPT ads.

Training. For Free, Plus, and Pro, OpenAI says "we may use your content to train our models" unless you turn off "Improve the model for everyone" (OpenAI).

Refusals. OpenAI has said it wants to treat "adults like adults," but in March 2026 it delayed ChatGPT's adult mode again with no new date (TechCrunch), and we found no later launch.

What each one still refuses, keeps, and sells against

This is the table we built the ranking from. "What its rules still bar" quotes each company's own usage policy: what the model is told to refuse. We didn't test refusals.

Switching from ChatGPT to Claude, Gemini, Perplexity, or Duck.ai fixes ads or training, but every one of them puts a ban on sexually explicit output in writing. Anthropic, Mistral, and Perplexity also put limits on medical and legal advice in their rules.

1. Morse

ChatGPT's three problems share a root. Your chats sit on OpenAI's servers, tied to your account, and the free plan may show ads. Morse keeps no chats on its servers, so there's nothing to train on and nothing to target an ad with, and its paid plans include models that aren't filtered the way the big names are.

• No ads on any plan. Morse shows no ads, including on Free, doesn't sell data, and runs no advertising trackers or third-party analytics. Nothing reads your conversation to pick an ad.
• Unfiltered models on paid plans. On paid plans, picking an unfiltered model gets answers other apps refuse: dark fiction, straight medical and legal answers, and help testing your own security. The Free plan's models aren't unfiltered. Morse never helps with anything illegal, and it reports CSAM to NCMEC.
• E2EE, for chats no one else should read. Your browser checks the enclave's hardware attestation, then encrypts the message so only that enclave can read it. Our 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 chats come with Pro, at $19 a month.
• Anonymous, for GPT, Claude, and Gemini without the provider knowing it's you. Our proxy removes your account, email, and IP address and sends the message under Morse's own account. The provider still reads the text and may keep it for a limited time to check for abuse, under its own policy, so leave your name out of what you write.
• Private, the default for new chats. Open-weight models and xAI's Grok run under zero data retention agreements Morse has with its inference, search, and other providers, so they keep nothing.

In every mode, your history is stored in your browser, encrypted, not on our servers, and Morse doesn't train AI on it. Because we don't store chats, we usually have very little to give authorities. A cleared browser takes the chats with it and we can't recover them, so save a copy of anything you need. Free is $0 with 15 messages a day across GPT-6 Luna, Claude Haiku 5.5, GLM 5.3 Flash, and more; Pro is $19 a month ($17 billed yearly) and adds E2EE chats and every unfiltered model.

2. Proton Lumo

Lumo has the strongest written privacy of the big names. Proton says "Lumo doesn't use your chats to train the AI models," that after a reply "the data is erased," and that saved history uses "zero-access encryption" (Proton). Its terms bar illegal activity and CSAM and say output isn't "a substitute for professional advice," with no separate ban on adult themes (Proton). Proton is also frank about the limit: "we need to provide the LLM with cleartext messages" (Proton). It offers no GPT or Claude models. The Free plan needs "No credit card," and Proton says you can start chatting "without even creating a Proton account"; Plus is $12.99 a month, or $9.99 billed yearly (Proton). We compare it with Duck.ai and Brave Leo in Duck.ai vs Lumo vs Brave Leo.

3. Duck.ai

Duck.ai is the easiest switch if you want GPT and Claude without handing over your IP address. DuckDuckGo removes "all metadata that contains personal information (for example, your IP address)," bars providers from "using Prompts and Outputs to develop or improve their models," and requires them to delete what they get "at most within 30 days, with limited exceptions for safety and legal compliance" (terms dated October 1, 2026, DuckDuckGo). Free models include GPT-6 Luna, Claude 4.5 Haiku, and Mistral Small 4; the Plus and Pro subscriptions add larger GPT and Claude models (DuckDuckGo). The same terms bar "Sexually explicit or obscene" content, so the refusals you know from ChatGPT come along.

4. Claude

Claude is the big-name pick for no ads. Anthropic wrote on February 4, 2026 that "Claude will remain ad-free" and that users "won't see 'sponsored' links adjacent to their conversations," adding it would be transparent if that ever changed (Anthropic). Training follows Settings, Privacy, "Help Improve our AI models" (Anthropic); chats used for training can be kept "for up to 5 years" de-identified, and flagged chats up to 2 years (Anthropic). Incognito chats, behind the ghost icon, are "not used for training" and kept 30 days by default (Anthropic).

Claude's written rules are strict in places. Its usage policy, in the version that takes effect November 12, 2026 according to Anthropic's page, bars "erotic chats" and content "related to sexual fetishes or fantasies," and requires a qualified human to review any "High-risk AI Recommendation," such as "diagnosing a condition" or "advising an individual on their legal rights" (Anthropic). Pro is $20 a month, or $17 a month billed yearly (Claude). More in is Claude private.

5. Le Chat, now Vibe

Mistral renamed Le Chat to Vibe on August 12, 2026, with "Same URL, same login" (Mistral). For Vibe, "Input and output data are used by default to train our artificial intelligence models"; on mobile you opt out under Data & Account Controls by clearing "Enable data sharing" (Mistral). Mistral keeps your chats "until you delete your account or until you delete the conversation," and says it hasn't engaged in "targeted advertising" in the preceding 12 months (privacy policy effective September 3, 2026, Mistral). Its usage policy, effective June 11, 2026, bars giving professional advice "without proper qualification," including "any form of health-related guidance" (Mistral). Pro is $14.99 a month, or $5.99 for verified students (Mistral).

6. Gemini

Google says that "Your Gemini Apps chats are not being used to show you ads," and that "If this changes, we will clearly communicate it to you" (Google). Privacy is weaker. With Keep Activity on, Google uses your chats to train its models; activity auto-deletes after 18 months by default; human reviewers read a subset, and "Reviewed chats are retained for up to three years." Even with Keep Activity off, chats are kept 72 hours. Google's guidelines bar "explicit or graphic sexual acts" and "gratuitous violence, whether real or fictional," so dark fiction can be refused too (Google). Google AI Plus is $4.99 a month and AI Pro $19.99 (Google). More in is Gemini private.

7. Perplexity

Perplexity is strongest at cited answers and weakest on this page's three tests. "For Free, Perplexity Pro and Perplexity Max users, AI Data Retention is enabled by default," and "Previously collected training data cannot be deleted or removed" (Perplexity). Its privacy notice, last updated October 19, 2026, says it may "share data with third parties to measure the performance of our advertising" and that "third parties may use this data to serve ads tailored to your interests" (Perplexity). Its acceptable use policy bars "pornographic or exploitative content" and restricts medical, legal, and financial services that need credentials (Perplexity). More in is Perplexity private.

How we compared them

We read every page linked above on October 20, 2026, except OpenAI's pages, which we read on October 19, 2026. We ranked on written policy only: what each company says it keeps, whether its pages describe ads or ad measurement, and what its usage rules bar. We didn't test how often any model refuses in practice.

Choose ChatGPT or Morse

• Choose ChatGPT if you rely on its native apps, voice mode, or custom GPTs, and you're happy to pay for Plus to drop the ads and switch off training yourself.
• Choose Morse if you want no ads on any plan, no training to opt out of, history kept encrypted on your device instead of a company's servers, GPT and Claude without the provider knowing it's you, and unfiltered models on a paid plan.

Ask your next question from the Morse home page without an account, or start free and open an E2EE chat today.

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

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