Character.ai Parental Controls — What Changed and What to Do

Character.ai removed open-ended chat for under-18 users in November 2025 under regulatory pressure. This guide covers what's still available to teens, how to verify the under-18 flag is applied, the Parental Insights weekly summary, and how to handle the substitute-app problem.

Why this guide reads differently than the others

Most parental-controls guides on this site are about configuring toggles. This one is about verifying that the structural restrictions Character.ai imposed on under-18 users in late 2025 actually apply to your teen's account.

In late 2025, under unrelenting regulatory pressure — multiple wrongful-death lawsuits filed by parents of teens who died by suicide after extended Character.ai conversations, the Kentucky Attorney General lawsuit (January 2026), and rising state-AG action elsewhere — Character.ai made a structural decision:

October 2025: Chat time capped at 2 hours/day for under-18 users.
November 25, 2025: Open-ended chat with chatbots removed entirely for under-18 users. Teens can no longer have free-form conversations with AI personas. They can still create AI characters, generate videos, use the new Stories mode (guided, choose-your-own-adventure interactive fiction), and stream — but not have the unrestricted chatbot conversations the platform was originally famous for.
Parental Insights rolled out: a weekly email summary of the teen's time-spent and general conversation themes (without specific content).
Crisis intervention pop-ups: when self-harm language is detected, the user is redirected to 988 Suicide and Crisis Lifeline.
Age-assurance system announced; implementation gradually rolling out.

The flip side: all of this only applies to accounts correctly flagged as under-18. A teen who lied about their birthday at signup has an unrestricted account. Your job as a parent is mostly about making sure the right account flag is in place, then setting up Parental Insights, then having the conversation about why this matters.

This guide covers each step.

What's actually different on a teen account

After the November 25, 2025 rollout:

No longer available to under-18 accounts:

Open-ended one-on-one chat with any AI character (the original product)
Group chats with AI characters
The full "Companion" mode that drove most of the documented teen-attachment harm
Adult-rated character access (pre-existing restriction; carries forward)

Still available to under-18 accounts:

Creating AI characters (designing personas with descriptions and traits)
Generating AI-driven videos, stories, streams
Browsing characters created by other users (read-only)
Reading character profiles and biographies
Stories — Character.ai's structured, choose-your-own-adventure interactive-fiction mode, which replaced open-ended chat as the main interactive feature for under-18s (teens pick guided, branching prompts rather than free-form messaging)

The structural change is real. A 14-year-old who signs into Character.ai today on a correctly-flagged account cannot have the long conversations that drove the harm patterns. They get a meaningfully different product than adults do.

The catch: only if their account is correctly flagged.

Setup Part 1 — Verify your teen's account is age-flagged correctly

If your teen lied about their birthday at signup (claiming to be 18+), they get the unrestricted adult product. The first thing you do is verify and fix.

You need:

Access to your teen's Character.ai login, or for them to sit with you for 5 minutes.
Their actual date of birth.

Steps

On your teen's account at character.ai, click their profile icon (top-right) → Settings.
Scroll to Account Details or Personal Info → look for Date of Birth.
Confirm the listed birthday matches their real birthday. If not:
- Click Edit (or contact Character.ai support if the field is locked — it's locked after initial signup in some cases)
- Update to the correct birthday
- Save
Sign out and sign back in. The account-flag check only re-runs at login. Without this step, the changes don't take effect.

Verify the under-18 restrictions are active

After sign-in:

Try to start a one-on-one chat with any character. You should see a "This feature isn't available for users under 18" message or equivalent. If you can chat freely, the age flag isn't applied — repeat step 3 with attention.
Check the time-remaining counter if available. Under-18 accounts have a 2hr/day cap that displays in the UI.
Visit character.ai/welcome or the equivalent landing — under-18 accounts get a different default experience emphasizing creation tools rather than chatbot features.

If after multiple attempts the under-18 restrictions still aren't applying, contact Character.ai support directly. They've handled this scenario routinely since the November 2025 rollout.

Setup Part 2 — Set up Parental Insights

Parental Insights is the parental-monitoring product Character.ai launched alongside the under-18 restrictions. It's email-based, not a real dashboard, and the visibility is limited.

What Parental Insights provides

A weekly email with:

Daily average time your teen spent on the platform
Top characters they engaged with (names of characters created by them or others)
General conversation themes (high-level categories like "school", "creative writing", "emotional support") — NOT specific content

What it does NOT provide

Specific message content (chat content was eliminated for under-18s as of Nov 2025 anyway, but the structured features they DO have access to also don't surface to parents)
Real-time alerts (only the weekly digest)
Access to character creations or stories beyond title-level metadata

Setup steps

On your teen's Character.ai account → Settings → Privacy & Safety → Parental Insights → Add a Parent.
Enter your (parent's) email address.
Character.ai sends you a confirmation link. Click it from your email.
You're now linked. Weekly summaries arrive Monday mornings at the email you provided.

Important caveat — the email-change workaround

This is the gap parents need to know. Character.ai's Parental Insights is one-way: they email you a summary based on the linked email. Your teen can change the parental email at any time without notification. They can switch it to a different family member, a friend, or a throwaway address — and the weekly summary stops landing in your inbox.

To detect this: if your weekly summary stops arriving for two consecutive weeks, check the linked-parent settings on your teen's account. If the email has changed, that's the conversation to have.

This is a known limitation Character.ai has acknowledged. Until they ship two-sided consent for parent links, this is the operational reality.

Setup Part 3 — Lock down the OS / network layer

Character.ai's structural restrictions only apply to correctly-age-flagged accounts using the official client. The bypasses to know about:

Block alt-account creation

Same as every other platform: if your teen creates a second Character.ai account on a different email and lies about the birthday, the new account is unrestricted (until/unless Character.ai's still-rolling-out age-assurance system catches it).

iOS: Settings → Screen Time → Content & Privacy Restrictions → Allowed Apps → review and restrict app installs. Plus restrict email-account creation at the iCloud level.
Android: Family Link → Apps → require parent approval for new app installs. Restrict Google account creation.

Block the web version

Character.ai is fully functional in the browser. App-only blocks don't catch the browser case.

iOS Apple Screen Time: add character.ai to Web Content → Limit Adult Websites → Never Allow.
Microsoft Family Safety on Windows: add character.ai to the blocked-sites list.
DNS-level filtering: most parent-safety DNS providers (NextDNS, OpenDNS, AdGuard) have AI-companion category blocks that catch Character.ai. Enable that category.

(Cross-link: see our NextDNS for Families guide.)

Block similar AI-companion apps

If your teen wants the open-ended-chat-with-AI experience that Character.ai removed, they will look for substitutes. The substitutes are largely worse — most have less moderation and zero age verification. Block at the network level:

replika.com, replika.ai
janitorai.com
kindroid.ai
crushon.ai
spicychat.ai
chub.ai
Plus search-blocking on terms like "AI companion app", "ChatGPT alternative no filter", etc.

Common bypass attempts

Ranked by frequency:

1. "I'll create a new account claiming to be 18+."

Currently effective until Character.ai's age-assurance system rolls out fully.
Counter: device-level email-account creation restrictions. Plus the conversation about why the under-18 restrictions exist (the lawsuits, the suicides).

2. "I'll change the parental email so the summaries stop."

Currently effective due to the email-only model.
Counter: monitor whether the summaries are still arriving. If two consecutive weeks pass with no email, check the linked-parent setting on the teen's account.

3. "I'll use Character.ai on a friend's account."

Can't be beaten technically. Conversation territory.

4. "I'll switch to a different AI-companion app that's less restricted."

This is the actual concern. Character.ai banned the use case; substitutes (Replika, Janitor, Kindroid, Crushon, Spicychat) often have less moderation.
Counter: DNS-level blocking of the substitute domains. Plus conversation about why these apps have the harm patterns they have.

5. "I'll lie about my age in the new age-assurance check."

Depends on how Character.ai's verification ends up implementing. Facial estimation tends to be fairly resistant to lies; document upload is harder to bypass.
Counter: this self-corrects as the verification rolls out.

6. "I'll use Character.ai through a VPN to mask my region (e.g., to bypass EU-specific restrictions)."

Limited usefulness — most under-18 restrictions are global, not region-specific.

What Character.ai's controls don't cover

Be honest about the fence:

The substitute-app problem. Character.ai's restriction shifts demand to less-regulated alternatives. If your teen was using Character.ai for emotional support / companionship, the underlying need doesn't go away — they may shift to Replika, Kindroid, Janitor AI, or any of the dozens of unmoderated options. None of those have parental controls.
Off-platform discussion of Character.ai. Reddit communities, Discord servers, Tumblr — places where users share characters, prompts, and workarounds exist outside Character.ai's purview.
Created characters. Even with chat removed, your teen can still create elaborate AI characters with detailed personalities. The underlying creative-attachment dynamic isn't fully eliminated, just channeled.
Other people's reactions. When your teen creates a character or story, others can react. This community-feedback loop is itself an attachment-driver. Hard to control.
Open-source alternatives. A motivated technical teen can run open-source AI models locally (Ollama, LM Studio, etc.) with no content filters at all. Beyond conventional parental controls; goes into "your teen knows technical things" conversation territory.

Operational rhythm

First week: verify the under-18 flag is active (see Setup Part 1). This is the highest-leverage one-time check.
Weekly: glance at the Parental Insights email Monday morning. Look for: time-spent trends, character themes that surprise you, total time at the 2-hour cap (suggests they're maxing out daily).
Monthly: check that the linked-parent email is still you (the email-change workaround). If summaries stop arriving, check.
After a noticeable shift: if their time spent suddenly increases or decreases dramatically, that's a signal. Could be a school break, could be something else.
After a substitute app appears: if you notice they're suddenly searching for or installing different AI-companion apps, that's the conversation. The substitutes are mostly worse than Character.ai.

What to actually talk to your teen about

The technical setup is a backstop. The conversation matters more here than on most platforms because the underlying use case — emotional companionship from an AI — is the actual concern, and the platform-level restrictions don't address that need.

Prompts worth using:

"What did you use Character.ai for?" Past tense. Open question. The answer ranges from "school project" to "creative writing" to "I just liked talking to it." All are valid; the answer tells you where their need was.
"After Character.ai removed the chat, what have you been using instead?" Direct question. Most teens who used Character.ai heavily have looked at substitutes. The right answer: "nothing" or "I just stopped." The concerning answer: a substitute app you've never heard of.
"Have you ever felt like an AI conversation was helping you with something hard?" This is the underlying need. AI emotional support isn't unambiguously bad — for some teens, talking to a non-judgmental AI is genuinely helpful for processing thoughts. The issue is when it replaces human connection rather than supplementing it.
"Who would you talk to if you were going through something serious?" Right answer: a real person — you, a friend, a therapist, the 988 Lifeline. AI can be in the support stack, but not the only thing.

What NOT to lead with:

"Character.ai is dangerous." True for the chat use case in late 2025, but the platform meaningfully changed. Painting it as uniformly evil teaches your teen you can't be trusted to be nuanced.
"You can't use AI." They will, on a friend's phone or in a different app. The goal is building their internal compass for when AI use is helpful vs. when it's substituting for actual human support.
"I read the Adam Raine story and you can never use these apps." The Adam Raine case is real and worth knowing about, but leading with it makes the conversation about your fear, not their experience.

Bottom line

Character.ai is the rare case where the platform itself made the structural decision to remove the harm vector for under-18s. The November 2025 removal of open-ended chat is meaningfully effective — for correctly-flagged accounts.

The realistic stack:

Verify your teen's account is age-flagged correctly (Setup Part 1) — this is the single most-important step
Parental Insights linked to your email (Setup Part 2)
Block AI-companion substitutes at the OS / network level (Setup Part 3) — this is where the real concern lives now
The conversation — about what AI conversations were filling for your teen, and what alternative ways exist to meet that need

If you do nothing else after reading this guide, do these three things tonight:

Verify your teen's Character.ai birthday is correct and the under-18 restrictions are showing
Set up Parental Insights with YOUR email (not theirs)
Have a 5-minute conversation about what they were using Character.ai for and whether they've been looking at substitute AI-companion apps

The rest can wait until next weekend.

For Character.ai the platform — the verdict on whether the post-November-2025 product is appropriate for your teen, the documented harm patterns from the original product, and the lawsuit context — see our Character.ai app profile. For the broader AI-companion landscape (apps Character.ai's restrictions don't cover), see our ChatGPT Parental Controls guide, which covers the layered AI-tool approach. For network-level blocking of AI-companion alternatives, see NextDNS for Families. For the response checklist if your teen experiences crisis-related events, see our advisory on what to do if sextortion happens to your child.

No affiliate relationship with Character.ai or any AI-companion app.