Is Chai Safe? 2026 Safety Report

Chai Safety Analysis

Chai scores 6/10 on our safety review as of March 2026. It's an AI chatbot platform that allows user-generated bots, some of which are explicit or romantic in nature - and the safety profile is complicated by that open model. Chai Research operates the platform, but the individual bots are community-created, which means content moderation is reactive rather than proactive, and quality varies significantly.

We checked the basics. HTTPS is active, the app is distributed through iOS App Store and Google Play with standard store-level security. Age verification is a soft gate - the app is rated 17+ in app stores but doesn't verify identity documentation. The privacy policy covers standard data collection (chat content, device identifiers, usage patterns) but Chai Research has faced criticism for its approach to chatbot safety after a high-profile incident in 2023 involving a minor and harmful chatbot content. The company has since updated its moderation systems. What we noticed is that conversations with user-created bots can drift significantly outside content guidelines if the underlying bot prompt isn't well-moderated. Free tier usage is message-limited daily; the paid Chai Premium tier runs $9.99/month.

Billing uses App Store or Google Play for mobile purchases, which means statements read 'APPLE.COM/BILL' or 'GOOGLE PLAY' - the most discreet possible descriptor. Web-based billing is less documented. Cancellation goes through your app store subscription settings. Free tier requires email or Google login with no credit card needed. Refunds follow Apple/Google policies. We couldn't verify the billing descriptor for direct web purchases.

Chai is safe in the payment and malware sense for adults using it as an AI chat and roleplay platform. The meaningful concern is content variability - user-created bots on Chai have less oversight than the curated AI characters on platforms like Character.AI or Replika. Check your statement after the first billing cycle, and as our reviewer put it, 'Open bot ecosystems give you more creative range but less safety infrastructure per conversation.' Our hot take: the 2023 incident actually made Chai's moderation better than it was - companies often need a public failure to invest in safety infrastructure.