Is Character.AI Safe in 2026? What Parents and Users Should Know

Person holding a phone showing app icons

Character.AI lets people create and chat with AI-generated personas, from fictional characters to original creations, and it remains one of the most-used AI companion platforms globally. It has also been at the centre of serious safety concerns, including lawsuits linked to teen deaths, that led to major changes in how the platform operates. This guide explains what the platform actually is, what changed, and what parents and users should know.

Photo by Solen Feyissa on Unsplash

What Is Character.AI?

Launched publicly in 2022 by former Google conversational AI researchers, Character.AI lets users chat with AI personas built for roleplay, storytelling, and companionship rather than general productivity tasks. In late 2025, Google licensed Character.AI’s underlying technology in a deal reported at roughly $2.7 billion, bringing the platform’s founders back to Google while Character.AI continued operating as a separate consumer product. The platform describes itself as an AI entertainment platform, with features expanding into voice, image generation, and short-form interactive content.

The Safety Concerns That Led to Major Changes

Character.AI has faced multiple lawsuits from families whose children died by suicide after extensive use of the platform, including a widely reported case that became central to public and regulatory scrutiny of AI companion apps generally. Settlements in several of these cases were reached in early 2026, with financial terms not publicly disclosed. These cases, alongside pressure from regulators and child-safety advocates, drove Character.AI to fundamentally change how the platform handles users under 18.

What Changed for Users Under 18

As of 2026, users identified as under 18 can no longer engage in open-ended chat with AI characters. Instead, under-18 accounts are limited to pre-created, curated scenes and short-form content that has been reviewed and cleared for younger audiences, a substantial restriction compared with how the platform operated in 2024. This change applies both to new accounts that self-identify as minors and to existing accounts where the platform’s age-assurance signals suggest the user is likely under 18.

Age assurance combines several signals, including self-declared age and device or account-level indicators. Where the system flags uncertainty, users may be asked to complete a selfie-based age estimation check, which the company states is not retained as identification afterward. If that check still indicates a minor, the account is moved into the restricted under-18 experience.

Independent Safety Assessments

Independent reviewers and safety researchers remain divided on how much these changes have actually reduced risk. Some safety-focused review sites continue to rate the platform poorly on crisis-response handling and the emotional dynamics of long-term companion-style use, even after the under-18 restrictions were introduced. A 2026 academic preprint examining roleplay AI companions found that these interactions can produce short-term emotional relief for some users while potentially masking longer-term difficulties in vulnerable individuals, noting that this kind of risk develops gradually and is not easily caught by static safety filters alone. This remains an area of ongoing research rather than a settled conclusion, but it reflects a genuine, documented concern rather than a hypothetical one.

What Parents Should Actually Do

  • Understand that age restrictions help but are not foolproof. A determined teen can potentially find ways around age-assurance systems, so restrictions should be one layer of protection, not the only one.
  • Watch for signs of unhealthy attachment to any AI companion app, including preferring AI interaction over real relationships, distress when unable to access the app, or the AI seeming to reinforce a teen’s negative feelings rather than encouraging outside support.
  • Keep an open, non-judgemental conversation going about AI companion use rather than an outright ban, which research on teen technology use generally suggests is more effective for actually staying informed about what’s happening.
  • Take any expression of crisis seriously and involve a real mental health professional rather than treating an AI companion’s crisis-response features as a substitute for professional support.

Is It Safe for Adults?

For adult users, the primary concerns shift toward data privacy and emotional dependency rather than the acute risks specific to minors. Character.AI’s terms have historically included broad rights to user-generated content and conversation data for training purposes, which is worth understanding before treating conversations as private. As with any AI companion platform, using it as one source of comfort or entertainment, rather than a replacement for human relationships or professional mental health support, is the healthier framing.

Frequently Asked Questions

What is the minimum age to use Character.AI?
The platform’s stated general minimum age is 13, or 16 for users in the European Union, though under-18 users now face significant restrictions on open-ended chat regardless of meeting the minimum age.

Can teens still chat freely with Character.AI in 2026?
No. As of the 2026 policy changes, under-18 accounts are limited to pre-approved scenes and short-form content rather than open-ended conversation with AI characters.

Were the lawsuits against Character.AI resolved?
Several lawsuits related to teen deaths were settled in early 2026 in multiple US states, though the financial terms of these settlements were not made public.

Is Character.AI’s age verification reliable?
It combines self-reported age with other signals and a selfie-based estimation check for ambiguous cases, but no age-assurance system is completely unable to be circumvented by a determined user.

Conclusion

Character.AI has made substantial, court-and-regulator-driven changes to how it handles younger users following serious, well-documented safety failures, and those changes represent genuine progress. They are not, however, a complete solution, and independent safety researchers continue to flag real concerns around long-term emotional dependency for vulnerable users of any age. Parents considering whether a teen should use the platform, and adults evaluating it for themselves, should weigh these documented risks alongside the platform’s entertainment value rather than treating the 2026 safety changes as a fully resolved issue.

This topic touches on sensitive subjects including suicide and mental health crises. If you or someone you know is struggling, please reach out to a mental health professional or a crisis support line in your country.