Frequently Asked Questions

Is toxicornot.ai free?

Yes. toxicornot.ai is completely free to use, with no account or signup required. Paste a message and you get an instant read.

Is it private? Do you store my messages?

When you paste a message for analysis, it is sent to our AI provider (Anthropic) to be processed in real time. We do not permanently store the content of the messages you submit — they are not retained after the analysis is complete.

As a precaution, please don't paste messages that contain passwords, financial details, or government ID numbers. You can read the full details on our Privacy Policy page.

How does it work?

You paste a message — a text, DM, email, or any snippet of a conversation — and the AI reads it for patterns associated with toxic or manipulative communication, such as gaslighting, guilt-tripping, love bombing, and stonewalling. It then explains what it noticed in plain language.

The analysis is powered by Anthropic's AI. It reads the words you give it; it has no other context about the people or the relationship beyond what you paste.

Can the AI be wrong?

Yes. The tool is meant to give you perspective, not a verdict. Tone, sarcasm, inside jokes, and missing context can all be misread — by an AI or by a person. Treat the result as one informed second opinion, not proof.

A single message rarely tells the whole story. Real toxicity shows up as a pattern over time, so use the read as a starting point for your own judgment rather than the final word.

Is this therapy or professional advice?

No. toxicornot.ai is an informational tool and is not a substitute for advice from a licensed therapist, counselor, doctor, or lawyer. It cannot diagnose anyone or tell you what to do.

If you're dealing with something serious — abuse, fear for your safety, or a real impact on your wellbeing — please reach out to a professional or one of the lines on our Help & Resources page.

What kinds of messages can I analyze?

Anything text-based: a single text or DM, an email, a group-chat exchange, or a back-and-forth you've copied out. The more complete the exchange, the more context the AI has to work with.

Does one toxic result mean my relationship is toxic?

Not on its own. One sharp message can come from a good person having a bad day, and one calm message doesn't clear a long pattern. What matters is whether the behavior repeats and how conflict is handled overall.

Our Guides go deeper on telling a rough patch from a genuine pattern, and on what healthy versus toxic communication actually looks like.

What if I'm in danger or in crisis?

toxicornot.ai is not emergency help. If you are in immediate danger, contact your local emergency services (911 in the US). For confidential support, the National Domestic Violence Hotline is available at 1-800-799-7233, or text START to 88788.

More hotlines — including international ones — and guidance on finding a therapist are on our Help & Resources page.

Who is toxicornot.ai for?

Anyone who has read a message and thought "wait, is it me, or was that not okay?" It's for the moment of doubt — when you want a calm, outside read before you decide how to respond.