Acceptable Use Policy

1. Illegal and harmful content

Do not use Calvyn to create, store, or deliver content that is illegal where you or your users are, or that:

• sexually exploits or endangers children, in any form, real or generated — we report this to the authorities and terminate immediately, with no notice and no refund
• harasses, threatens, bullies, or incites violence against a person or group
• promotes self-harm, suicide, or eating disorders
• incites or facilitates terrorism or violent extremism
• is defamatory, or that you know to be false and damaging to someone
• infringes copyright, trademark, trade secret, or another intellectual property right
• discloses someone’s private or identifying information without their consent
• is sexually explicit material, or content that is obscene under applicable law

2. Deception and impersonation

Do not configure an agent to claim it is a human being where that would deceive the person talking to it, and do not deny that it is an AI if asked directly. Several jurisdictions require disclosure; beyond the law, it is the thing that most reliably destroys a user’s trust in your business.

Do not impersonate another person, business, or brand you are not authorised to represent, and do not use Calvyn to imply an endorsement or affiliation that does not exist.

Do not use Calvyn to run phishing, romance scams, fake support desks, fraudulent investment pitches, or any other scheme designed to trick people out of money or credentials.

Do not generate content designed to manipulate an election or mislead people about voting.

3. High-risk uses

AI output can be wrong in ways that read as confident and correct. Do not deploy a Calvyn agent as the sole decision-maker, without a qualified human reviewing its output, in any of the following:

• medical diagnosis, treatment, medication dosing, or triage — human or veterinary
• legal advice, or anything a user could reasonably act on as legal advice
• individualised financial, tax, or investment advice
• decisions about a person’s employment, credit, housing, insurance, or education
• emergency, crisis, or safety-of-life situations, including any use where a wrong answer delays someone getting real help
• operation or monitoring of critical infrastructure, vehicles, industrial systems, or medical devices

If your agent could plausibly receive a message from someone in crisis, configure it to hand off to a human or to direct the person to an appropriate emergency service.

4. Privacy and other people’s data

Only upload knowledge you have the right to use. Do not upload someone else’s confidential material, or personal data you have no lawful basis to process.

Do not configure agents to solicit sensitive personal information from end users — health conditions, biometric data, precise location, government identifiers, or full financial account details.

Do not use Calvyn to build profiles of individuals, to conduct surveillance, or to scrape and reassemble personal data about people who have not consented.

If you deploy an agent to users in a jurisdiction with data protection law, you are the controller of those conversations and it is your obligation to give those users notice and a lawful basis. Our Data Processing Addendum covers our side of that.

5. Protecting the service

Do not attempt to breach, probe, or circumvent Calvyn’s security, authentication, rate limits, or plan allowances.

Do not access another customer’s account, agents, or conversations.

Do not scrape the service, reverse-engineer it, or use it to build a competing product from our own output.

Do not send malware, run denial-of-service traffic, or use Calvyn as a relay to attack a third party.

Do not use Calvyn to send unsolicited bulk messages, or in any way that breaches anti-spam law such as the CAN-SPAM Act.

Do not resell, sublicense, or share your account access outside your own organisation unless we have agreed to it in writing.

Automated or programmatic use must stay within your plan’s documented limits. If you need more, talk to us rather than working around them.

6. How we enforce this

We investigate reports and act on what we find. Depending on severity we may warn you, require you to change a configuration, restrict a feature, suspend the account, or terminate it.

For most breaches we will tell you what the problem is and give you a reasonable chance to fix it. For content that sexually exploits children, for active attacks on the service or its users, and for anything creating immediate legal exposure or risk to a person’s safety, we act immediately without notice.

Termination for breach of this policy does not entitle you to a refund of fees already paid.

We are a small operation and we do not proactively read customer conversations. Enforcement is complaint-driven and incident-driven, which means the standards above are yours to uphold in the first instance.

7. Reporting abuse

If you believe a Calvyn agent is being used in breach of this policy, email hello@calvyn.io with the website address where you found it and what you saw. We read every report.

If you believe content infringes your copyright, send a notice with enough detail to identify the work and the material to legal@calvyn.io.