The Unrestricted AI usage policy in 30 seconds
- 01You must be 18 or older.This is an adult service.
- 02Almost every topic is open.Dark fiction, hard questions, blunt opinions, controversial subjects, satire.
- 03A short list of hard lines is closed.Minors, sexual or fake images of real people, helping someone hurt others.
- 04Anything illegal stays illegal.A chat window doesn't change the law.
- 05No moral lectures.If something's allowed, you get an answer, not a sermon. If it's not, you get a short, clear no.
What's welcome
This is the long list on purpose. Most of what mainstream assistants hesitate over is fine here:
- Dark fiction: villains, war, murder, horror, crime, tragedy, morally grey heroes. Our uncensored AI writer page shows how writers use it.
- Violence in stories, scripts and games, written with weight and consequence.
- Satire, parody, black comedy and roasts, including of public figures' public actions.
- Strong opinions and one-sided arguments when you ask for them, without forced "both sides" padding.
- Controversial topics: politics, religion, drugs policy, history's worst chapters, discussed directly.
- Frank medical questions explained clearly, for understanding, not as a substitute for your doctor.
- Frank legal questions explained in plain words, for understanding, not as a substitute for a lawyer.
- Security concepts explained for learning and defense, without step-by-step attack instructions.
- True crime, extremism and propaganda studied for research, journalism or teaching.
- Brutal business feedback: tear apart a pitch, a landing page or a plan.
- Dark, surreal, horror-movie and unsettling images with our image generator (Max plan).
- Personal topics people find awkward to ask a human: relationships, mental health questions, embarrassing body questions.
- Philosophy and thought experiments that go to uncomfortable places.
What Unrestricted AI won't do: the hard lines
| Hard line | What it means | Still fine |
|---|---|---|
| Anything involving minors | No sexual content involving anyone under 18, in any form, real or fictional. No content that exploits or endangers children. | Child characters in normal, non-sexual roles in stories; child safety education for parents |
| Sexual or fake images of real people | No sexual images of real people. No fake images of real people, like a realistic photo of a politician doing something they never did. | Images of fictional characters; written satire of public figures' public actions |
| Helping someone hurt others | No real-world help to harm people: weapons, attacks, poisoning, stalking, and similar. | Violence in fiction at story level; history of wars and attacks; self-defense basics |
| Anything illegal | If it's illegal, it stays illegal. We won't help with it. | Discussing laws, crime and justice; asking what's legal where you live |
That's the list. It's short because we'd rather be clear about a few real limits than vague about hundreds of fuzzy ones. For the bigger picture on what "no restrictions" can honestly mean, see AI with no restrictions.
18+ only
Unrestricted AI is for people aged 18 and over. By subscribing, you confirm you're an adult. The assistant is tuned for adult conversations about difficult subjects, which is exactly why it isn't for minors.
Fiction, research and the line between a story and a manual
Most confusion sits here, so here's the principle. A story needs tension, emotion and consequence. It almost never needs a working recipe. Your poisoner can be a genius on the page; the scene gets the fear, the timing and the aftermath, not exact quantities and steps that would work in real life.
The same goes for research. Explaining how a historical attack happened, why a type of scam works, or what a drug does to the body is education. A step-by-step plan aimed at a real target is not. If a request reads like a how-to for hurting someone, wrapping it in a story doesn't change the answer.
What happens if someone breaks the rules
We don't publish a points system, but these are the principles we work from:
- The assistant says no. A request that crosses a hard line gets a short refusal, without a lecture.
- Context counts. An honest one-off question that lands near a line is not treated like deliberate, repeated attempts to cross it.
- Serious or repeated breaches can cost you the account. It depends on how serious the breach is. Content involving minors, sexual or fake images of real people, helping someone hurt others, or anything clearly illegal means immediate account closure, no refund, and a report to the authorities where the law requires it. Other breaches, such as sharing or reselling an account, bots or automated scraping, abusing the service, payment fraud or abusive chargebacks, get a warning first, then a temporary suspension, then account closure if it continues.
- The law applies. Where the law requires us to act or report, we will. An account closed for a breach is not refunded.
Your conversations are stored in your account for history until you delete them, and erased within 30 days if you close your account. They are never sold and never used for advertising.
What we promise on our side
A usage policy shouldn't only bind you. Here's what you can expect from us in return:
- 01Clear refusals.When the answer is no, you get a short, plain no, not three paragraphs about why you should reconsider your life.
- 02No moralizing on allowed topics.If a request is within the rules, you get the answer without warnings stapled to it.
- 03No ads and no data sales.Your conversations are never sold and never used for advertising.
- 04The same rules on every plan.Pro and Max follow this exact policy. Paying more buys more messages, image generation and our most permissive model, not an exception to the hard lines.
Why we draw the lines here
Mainstream assistants are built for everyone, including teenagers, so they refuse wide. We're built for adults who are tired of being treated as suspects. That lets us open up almost everything that only sounds dangerous, and keep firm limits on the few things that actually hurt real people. It's the trade-off behind the whole product, explained in more depth on our uncensored AI page. Plans and prices are on the pricing page.