Did Character.AI Remove Age Verification? The Rules in 2026

How this guide was made. This is a documentation-based guide. On 27 October 2026 we read Character.AI’s own blog posts and help centre articles, the processor privacy policy of its age verification provider, Persona, and the Character.AI listing on the US App Store. Dates in this guide come from those pages. One help article, “What is the age requirement?”, would not load for us, so we make no claim about it. Nothing here is hands-on testing and none of it is legal advice. Our method is set out on how we review.

The short version

Character.AI no longer lets users under 18 have open-ended chats with its characters. Teens can still use the platform, but in a separate experience built around reading, creating and other non-chat features.

To decide who goes where, Character.AI runs what it calls age assurance on every account. For most adults this happens quietly in the background. A selfie, and in some cases a government ID, only comes into it if the system thinks you might be under 18 and you want to show that you are not.

This guide explains what changed and when, how the checks work according to the company’s own pages, what happens to the data, and what to do if you are an adult who has been placed in the wrong experience.

What Character.AI announced

On 29 October 2025 Character.AI published a post titled “Taking Bold Steps to Keep Teen Users Safe on Character.AI”. It said it would remove the ability for under-18 users to engage in open-ended chat with AI, “no later than November 25”.

During the transition, under-18 chat was capped at two hours per day, with the limit ramping down before the deadline. The same post said the company had “built an age assurance model in-house” and would combine it with “leading third-party tools including Persona”. It also announced plans to establish and fund an independent non-profit called the AI Safety Lab.

A follow-up on 21 November 2025 set out the rollout. In the US, removal would begin on Monday 24 November 2025, with other markets “following in short order”. Users were split into two groups: the first lost open-ended chat from 24 November over several days, and the second weeks later, with at least two weeks’ notice. US teens were moved to a one-hour daily limit in the meantime. The post said age assurance was rolling out in the US first, with a global rollout “in the near future”.

On 3 September 2026 the company described the removal as “one of the most significant safety decisions we have ever made” and said it remains in place.

What teens can still use

Character.AI’s help centre calls the under-18 experience reading mode. According to its “What is Reading Mode?” article (updated 17 July 2026), teens can browse characters and view content, and they keep:

• all past chats, saved and viewable
• creating characters, scenes and videos
• Imagine Chat and AvatarFX
• Feed activity, such as sending stickers, reacting and supporting creators

The blog adds more. Stories, launched on 25 November 2025, is described as “a guided way to create and explore fiction, in lieu of open-ended chat”. The September 2026 post lists c.ai Series, Feed and Comics as formats being expanded for under-18 users.

For parents, the same post says Parental Insights was improved through a partnership with k-ID. Parents or guardians whose email is linked to a teen’s account can see the teen’s platform activity and receive a weekly email.

How the age checks work

The help centre describes a layered process. Here it is in order, using the company’s own wording where it matters.

Step 1: signals. Character.AI’s Age Assurance FAQ (updated 1 August 2026) says it runs age assurance technology on all accounts, looking at “a number of signals, including login info, activity on the platform, and some signals from third parties”. If that indicates you are 18 or over, “you’re good to go”. The September 2026 blog post says this rests on the company’s own “in-house age estimation model”.

Step 2: a selfie, only if flagged. If the signals suggest you are under 18, you are told. You can then stay in the under-18 experience or choose to verify. Verification is done by Persona using a selfie. The FAQ says you will not be asked for a selfie “unless we believe you’re under 18 and you want to prove that you’re actually 18+”.

Step 3: an ID, only as a last resort. If Persona “isn’t sure you’re 18+” from the selfie, it asks for an ID. The tips article (updated 11 September 2026) says the photo must be of a physical ID, not a digital copy, and lists accepted documents: driver’s licence, passport, passport card, identification card, permanent residence card, immigration visa, consular ID and certificate of citizenship. It also says there is “no alternative” to a selfie or government ID if you are prompted.

A few points the FAQ makes clear:

• The birthday you entered is not the deciding factor. A separate article says Character.AI asks for it to meet minimum age rules and legal obligations.
• A Character.AI+ subscription “does not bypass or override” age assurance.
• A Google account with an age on file does not bypass it either.
• If you are flagged and have no government ID, “the verification process cannot be completed”.

One wording point is worth noting. The “What is Age Assurance?” article says “All users are required to complete an age assurance check”, while the FAQ says most users will not need to do anything. Read together, the check itself covers everyone, but the selfie step applies only to accounts the system flags.

What happens to your data

According to “What happens to my data during age assurance?” (updated 14 July 2026), Character.AI “does not collect or store any personal data submitted during age verification”. Persona handles submissions, and “any biometric data or ID submitted is automatically and permanently deleted within 7 days”. You do not need to request deletion.

The FAQ also says Persona processes the information “to verify your age, identify fraud, and for quality assurance”. It notes that if Character.AI ever launches virtual currency or similar features, it may need “very limited exceptions” to access an ID through Persona for fraud issues, and says it would tell users and get consent before any material change.

Persona’s own Processor Privacy Policy (last updated 9 April 2026) says that for selfie-only age estimation it analyses “non-uniquely identifying facial geometry”, and that this data “is deleted immediately as soon as an outcome has been determined”. It says its default is to delete all personal data once processing is complete, and that it will not sell, lease or trade this data. The policy is written for people using Persona through a customer such as Character.AI, so the customer’s settings matter. Character.AI’s “within 7 days” is the figure it gives for its own setup.

If you are an adult wrongly flagged as under 18

This is the situation most adult users worry about. Users report being moved into reading mode unexpectedly, and the company’s September 2026 post says it is working to improve accuracy “so users who are actually 18 or older will have fewer disruptions”. Calm, practical steps:

1. Start verification from Settings. In the app or on character.ai, go to Settings, then Advanced, then Verify Age. If you cannot see it, update the app. The help centre says the feature rolls out gradually, so not everyone sees it at once.
2. Prepare for the selfie. The tips article advises good light on your face, the camera at eye level, no filters, masks or sunglasses, only one face in the frame and slow movement. It warns that filters or makeup effects that change how you look may cause a decline.
3. Use your best camera. You can switch devices mid-process using the QR code shown during verification.
4. Have a physical ID ready, in case the selfie is inconclusive.
5. Contact support if it goes wrong. If you were verified but think you are in the wrong experience, or you cannot attempt verification any more, the help centre says to submit a ticket through its support form so the team can review your age bracket or restart the process.

Teens who turn 18 can also move to the adult experience. The help centre says to contact support so it can reset verification attempts.

Dated timeline

What we could not confirm

• When age assurance reached each country outside the US. The November 2025 post promised a global rollout but gave no dates, and we found no official country list.
• When the second group of US teens lost open-ended chat. The company gave notice to those users but did not publish the date.
• How accurate the signal-based system is, or how often adults are flagged. No figures were published.
• Exactly which “third-party signals” are used.
• What the help article “What is the age requirement?” says; it would not load for us.
• Whether the AI Safety Lab has launched; the pages read for this guide do not say.

The US App Store listing, read the same day, is rated 18+ and lists “Age Assurance” and “Parental Controls” among its in-app controls.

For parents

If your teen uses Character.AI, the main change is simple: they can no longer chat freely with characters, but they can still browse, create and read. Ask them calmly what they use it for. Parental Insights may help if you want a weekly summary. Our age and content policy explains how Aitryvo covers these apps, and our privacy and safety guide compares what several apps say about data.

Adults weighing other options can see our Character.AI alternatives or browse the apps directory.