Is JanitorAI safe?

JanitorAI has several basics in place: an 18+ app rating, age checks in some countries, card payments handled by Stripe, self-serve account deletion and a written ban on deepfakes of identifiable people. The main gap is that its chat data policy was not available to check, so you cannot confirm how your chats are used. Here is what the official pages said on Oct 25, 2026, and what they did not.

The short answer

On paper, JanitorAI is a reasonably normal platform with a few unknowns. Nothing in the official pages I read is alarming. But “safe” depends partly on things JanitorAI did not publish where I could find them, mainly how chat data is used and where the company is based.

If you want to keep risk low, treat it like any chat service you do not fully know: do not put real names, addresses or anything identifying into chats, and be careful about which outside model you connect.

Who runs it

The App Store lists the developer as JanitorAI, Inc. Its country or jurisdiction is not stated on the pages checked.

That matters less for day-to-day use and more if something goes wrong. Knowing where a company is based tells you which privacy laws apply and who you would complain to. Here, that is an open question.

Age checks

JanitorAI’s app is rated 18+ on the App Store. Beyond that, the official help center describes a one-time age verification through a provider called k-ID, for users connecting from:

• Australia, from Apr 24, 2026
• Brazil, from Apr 24, 2026
• The UK, from Jun 15, 2026

You can verify by facial age estimation, an ID scan or AgeKey. JanitorAI says it receives only a pass or fail result, not your face scan or ID. That is a sensible design: the sensitive data stays with the verification provider.

For other countries, including the US and Canada, the verification method is not stated officially.

There is also a content setting: Limited vs Limitless. The help center says the NSFW toggle for the mobile app is set on the website.

Your data

Chats

How JanitorAI uses chat data, including whether chats train models and whether staff can read them, is not published on the pages I could check. The terms and privacy pages did not load for the research tools, so this is “not verified”, not “bad”. If this matters to you, read the privacy policy yourself before you start long or personal chats.

Outside models

If you connect your own model through a proxy and API key, your messages also go to that provider. The official proxy guide says costs and logging depend on the external provider. In other words, connecting a proxy adds a second company’s privacy policy to the one you already have to trust. Pick a provider whose policy you have read.

Payments

According to the help center, web payments are processed by Stripe, and JanitorAI says it does not store your card details. App subscriptions go through the App Store or Google Play, which means the app stores hold your payment details instead. Payments are generally non-refundable.

Deleting your account

You can delete your account yourself:

1. Go to Settings > Security > Delete account.
2. Confirm with the 6-digit code sent to your email.

Deletion is permanent. The help center says your posts, characters, personas and profile are removed. It does not explicitly mention chats. Support cannot delete accounts for you, so you have to do it from your own login.

If you pay through the App Store or Google Play, cancel the subscription there as well. The help center says app subscriptions cannot be cancelled from the web, and deleting an account is not the same as cancelling billing.

Content rules

JanitorAI’s harassment rules are clear about real people:

• No content that depicts or recreates real private individuals
• That includes AI likenesses and deepfakes of identifiable people
• Widely recognized celebrities may be depicted
• Content creators, influencers, streamers and sex workers may not be depicted in any way

That last group is a useful detail. Plenty of people who are well known online are still private individuals in most senses, and JanitorAI draws the line to protect them.

What users report

JanitorAI had a Trustpilot rating of 2.5/5 from 35 reviews at the time of checking, a small sample. The safety-related theme is moderation: reviewers report automated mutes and bans with unclear reasons, and slow support. Others mention repetitive replies, quality drops after model changes, removed features and chats cut off. Those are frustrations rather than safety risks, but an unexplained ban with no quick human answer is worth knowing about before you invest hours in characters. Reddit complaints were not collected.

What is not published

To be fair to JanitorAI, these are gaps in what I could verify, not proof of a problem:

• The company’s country or jurisdiction
• How chat data is used, including model training and staff access
• Whether chats are removed when you delete your account
• The age-verification method outside Australia, Brazil and the UK
• The web (Stripe) price for Janitor+

How to use it more safely

• Keep personal details out of chats and character descriptions.
• Read your proxy provider’s privacy policy before connecting it.
• Use a unique password, and sign up with an email you can always reach, since deletion needs a code sent there.
• If you pay in the app, remember that cancelling happens in the App Store or Google Play.
• If you are done with it, delete the account yourself from Settings > Security.

For prices, features and my overall view, see the JanitorAI test drive. If the moderation complaints put you off, there is a list of JanitorAI alternatives.