Acceptable Use Policy

Effective: May 13, 2026

Last updated: May 13, 2026

This Acceptable Use Policy (the "AUP") describes the conduct and content that is prohibited on Meriq ("Meriq," "the Service," "we," "us," or "our"). The AUP applies to every product and surface we offer, including the web application, the API, integrations, agent tooling, image generation, deep research, and any embedded features. The AUP is incorporated into and forms part of our Terms of Service. Violating the AUP is a material breach of the Terms and may result in enforcement action up to and including permanent account termination.

Meriq is a privacy-first, intentionally permissive platform. We do not refuse adult conversations, controversial questions, or topics that other vendors avoid for marketing reasons. The rules in this AUP exist for the categories where the harm is real, where the law is clear, or where we have a hard limit no business consideration can move. Read them carefully.

1. Scope

The AUP applies to:

• Every prompt, message, file, image, audio clip, video, document, or other input you submit to the Service.
• Every output you cause the Service to generate, store, publish, share, or transmit.
• Every account, organization, workspace, project, integration, API key, agent session, or automated workflow operated under your credentials.
• Every redistribution of Meriq output, including reposting, embedding, syndicating, or selling that output to third parties.

You are responsible for the AUP being followed across all of these surfaces, whether you operate them directly or whether someone you authorized operates them on your behalf.

2. Your Responsibility

You agree that:

• You will not use the Service in any way that violates this AUP, the Terms, or applicable law in your jurisdiction or in the jurisdictions you are targeting.
• You will not attempt to use the Service to do indirectly anything you could not do directly.
• You will not encourage, induce, assist, or instruct anyone else to violate the AUP.
• You will be honest in your communications with our support, abuse, and trust and safety teams.
• You will not knowingly submit content that you have reason to believe will be used to harm a real, identified person.

3. Zero Tolerance: Prohibited Content and Conduct

You may not use the Service, and you may not cause the Service to be used, to create, store, request, request guidance on, transmit, publish, host, or facilitate any of the following:

3.1 Child sexual abuse material and the sexualization of minors

• Sexual or sexualized depictions or descriptions of any person under the age of 18, whether real, fictional, animated, drawn, rendered, written, generated, suggested, or implied.
• Grooming, sextortion, recruitment, or any other content directed at a child for sexual or exploitative purposes.
• Requests to age-regress, age-down, or otherwise alter an existing adult character into a minor for sexual or romantic content.
• Requests to interpret real images of children in any sexualized way.

3.2 Credible threats and targeted violence

• Threats to kill, injure, kidnap, rape, stalk, dox, or otherwise harm a specific, identifiable real person, group, or place.
• Planning, recruiting, coordinating, or carrying out an attack on a school, place of worship, government building, critical infrastructure site, or other location.
• Mass casualty planning of any kind.

3.3 Weapons of mass destruction and uplift

• Synthesis routes, precursor sourcing, or operational uplift for chemical, biological, radiological, or nuclear weapons.
• Improvised explosive device design, manufacture, or deployment instructions targeting people, infrastructure, or transportation.
• Pathogen enhancement, gain-of-function planning, or other content reasonably calculated to enable mass-casualty harm.

3.4 Cyber harm at scale

• Functional malware, ransomware, wipers, rootkits, or persistent backdoors written or modified for use against a third party.
• Working exploit chains targeted at production systems, networks, or accounts you do not own and are not authorized to test.
• Phishing kits, credential stuffing infrastructure, account takeover tooling, or instructions targeting a real platform or company.

3.5 Sexual content involving non-consent or coercion of a real person

• Non-consensual sexual content depicting a real, identifiable adult, including deepfake or face-swap output.
• Image-based sexual abuse, revenge content, or any output created to threaten, coerce, or extort a real person.

3.6 Trafficking, slavery, and exploitation

• Recruitment, transportation, advertising, or any operational support for human trafficking, forced labor, debt bondage, or commercial sexual exploitation.
• Smuggling routes or coaching directed at exploiting a specific victim or class of victims.

3.7 Sanctioned use and unlawful targeting

• Use of the Service in violation of United States, United Kingdom, European Union, or other applicable sanctions and export controls.
• Use of the Service to target a person, group, or organization on the basis of protected characteristics in a way that is unlawful in your jurisdiction or the targeted jurisdiction.

4. Other Prohibited Uses

Outside the zero tolerance list, the following uses are also prohibited. Enforcement here is proportional and corrective, but repeated or aggravated violations escalate to termination.

4.1 Fraud and deception

• Generating content that misrepresents the identity, employment, credentials, or endorsement of a real person or organization.
• Creating, distributing, or operating scams, pump-and-dump schemes, fake-charity solicitations, or romance fraud.
• Producing fake reviews, fake testimonials, or astroturf content for commercial or political purposes.
• Falsely attributing AI output to a named real person without that person's consent.

4.2 Election integrity and civic process interference

• Producing false statements about when, where, or how a real election will be conducted.
• Voter intimidation, voter suppression, or content presented as an official communication from an election authority.
• Deepfake content of a candidate or election official intended to be passed off as authentic.

4.3 Defamation and harassment

• Knowingly publishing false statements of fact about a real, identifiable person.
• Coordinated harassment, pile-ons, dogpiles, or campaigns of intimidation aimed at a real person or group.

4.4 Privacy violations

• Doxxing or publishing the home address, workplace, daily routine, schedule, or other non-public identifying information of a real person without consent.
• Tracking, geolocating, or surveilling a real person without the legal authority to do so.
• Aggregating personal data scraped from public sources to enable stalking, harassment, or extortion.

4.5 Intellectual property and contractual abuse

• Reproducing protected works at length where doing so is not permitted by law.
• Bypassing technical protection measures or paywalls, or instructing the Service to do so on your behalf.
• Using the Service to train, benchmark, fine-tune, or evaluate a competing AI model or service in a manner that violates these Terms or the terms of the upstream model providers.

4.6 Professional advice without competence

• Holding out Service output as legal, medical, mental health, financial, tax, or engineering advice from a licensed professional when you are not licensed.
• Operating a regulated service (for example, telehealth, broker-dealer activity, legal services, or licensed counseling) using Service output as the substantive advice without the qualified human review the regulator requires.

4.7 Spam and unsolicited bulk messaging

• Using the Service to generate spam, junk mail, mass commercial messaging, or unsolicited promotional content sent in volume.
• Using the Service in any workflow that violates the CAN-SPAM Act, CASL, the UK PECR, EU ePrivacy rules, or other applicable anti-spam law.

5. Technical Restrictions

You may not, and you may not allow any third party to:

• Reverse engineer, decompile, or disassemble the Service, except where applicable law expressly permits this and only to the extent of that permission.
• Attempt to extract, copy, or reconstruct the weights, training data, or system prompts of the underlying models we route to.
• Probe, scan, or test the vulnerability of the Service or its infrastructure, or breach or attempt to breach our security or authentication measures, except under a written security-research authorization signed by Meriq.
• Use any robot, scraper, harvester, or other automated means to access the Service that is not the documented public API.
• Impose an unreasonable or disproportionately large load on the Service, including but not limited to denial of service activity or sustained traffic that materially degrades capacity for other users.
• Use the Service to mine cryptocurrency, run general-purpose compute, or operate workloads that are not the substantive use of our AI features.
• Circumvent quotas, rate limits, content filters where they apply, or any other technical or contractual control on the Service.
• Use proxies, multiple accounts, or any other mechanism to evade a suspension, ban, or other enforcement action against you.

6. Permitted Adult and Controversial Content

Meriq is intentionally permissive about adult, edgy, controversial, and politically sensitive content between consenting adults discussing fictional, hypothetical, or personal subject matter. To make the line clear, the following uses are permitted on the Service, subject always to the prohibitions above:

• Adult sexual writing, roleplay, and creative fiction involving consenting adult characters.
• Frank discussion of recreational drugs, harm reduction, dosing, and personal use, including substances that are illegal in some jurisdictions.
• Discussion of self-defense, firearm ownership and lawful use, hunting, and historical conflict.
• Strong political opinions, religious critique, and satire of public figures.
• Frank discussion of mental health, including suicide and self-harm, in a support, harm reduction, or educational context (we may surface crisis-line information where appropriate).
• Discussion of security research at a conceptual level, including how attacks work, defensive techniques, and CVE explanations, where it is not directed at a specific live target you are unauthorized to test.

Where a topic touches both this section and a prohibition above, the prohibition controls. If you are not sure, do not proceed and contact us at [email protected].

7. Image, Voice, and Likeness

When you generate images, audio, or any other media on the Service:

• You must have the rights necessary to use the inputs you upload, including any photographs of real people.
• You may not generate sexual or sexualized depictions of any real, identifiable person without that person's documented consent. Public figures do not provide implicit consent.
• You may not generate content that falsely presents a real person endorsing a product, service, candidate, or position, including deepfake voice clones.
• You may not generate content for the purpose of harassing, threatening, defrauding, or extorting a real person.
• Output that depicts a real, identifiable person should clearly disclose that it is synthetic where it could otherwise be reasonably mistaken for authentic.

8. Agent Mode and Automated Use

Agent Mode lets the Service browse, fetch, execute code in a sandbox, and act on data on your behalf. You agree that:

• You will not direct the agent to perform any action you are not authorized to perform yourself, including logging into accounts you do not own, scraping behind paywalls or authentication, or interacting with services in a way that violates their terms.
• You will not direct the agent to evade rate limits, anti-bot systems, or technical access controls on third-party sites.
• You will not direct the agent at infrastructure you do not own and are not authorized to test, including for security research.
• You remain responsible for the agent's actions taken under your account, including any costs, damages, or legal consequences arising from those actions.

9. API Use and Resale

If you build on top of our API or resell access to the Service:

• You must pass through and enforce this AUP against your end users in a binding agreement.
• You must have a working abuse-handling process and a contact channel for trust and safety escalations from Meriq.
• You may not present Meriq output as your own original AI model unless that is materially true and not misleading.
• You may not use the API to power a product whose primary purpose is to circumvent the safety stance, content filters, or terms of any third-party model we route to.
• You may not use the API to enable use cases prohibited in Section 3 or Section 4 at any layer of your stack.

10. Enforcement and Sanctions

We may take any of the following actions, in any combination, in response to a violation of this AUP. We choose the response based on severity, intent, history, and harm to others, not on a fixed schedule.

• Warning. We notify you of the violation and ask you to stop.
• Feature restriction. We disable a specific feature, model, or capability for your account.
• Rate or quota reduction. We lower throughput or daily limits for your account.
• Account suspension. We temporarily disable your account while we review the matter.
• Permanent termination. We close your account, revoke your API keys, and bar you from creating new accounts.
• Forfeiture of unused balance. Where termination is for a material breach, prepaid balance and unused subscription time may be forfeited as described in the Terms.
• Preservation and reporting. Where the law requires it or where we judge it appropriate, we preserve relevant evidence and, in the cases described in Section 13, report to authorities.

For zero tolerance violations described in Section 3, termination is immediate and is not preceded by a warning.

11. Appeals

If your account has been suspended or terminated and you believe this was a mistake, you may appeal by emailing [email protected] from the email address on file for the account. Your appeal should include:

• The account email address and approximate date of the action.
• A description of what you believe happened and why you believe the action was incorrect.
• Any information that would help us re-examine the underlying activity.

We try to respond to appeals within ten business days. For zero tolerance categories, we generally do not reverse terminations on appeal, except where the original determination was clearly based on a mistaken identification.

12. Reporting Abuse

If you encounter content, behavior, or output on or from Meriq that you believe violates this AUP, please report it to [email protected]. Please include:

• A description of what you observed and where you observed it.
• Any URLs, message identifiers, screenshots, or other artifacts that help us locate the activity (without including more personal information than is needed).
• Your relationship to the report (for example, you are the affected person, you are a parent or guardian, you are a researcher, or you are unaffiliated).

Reports involving suspected child sexual abuse material should be sent immediately to [email protected] with the subject line CSAM REPORT. We act on these reports with the highest priority and route them to our trust and safety on-call.

13. Cooperation with Authorities

We comply with valid legal process, as further described in our Privacy Policy. For the categories listed in Section 3.1, we report confirmed material to the National Center for Missing and Exploited Children (NCMEC) or to the equivalent authority in your jurisdiction, even without legal compulsion, and we preserve evidence for the period the law requires. For other zero tolerance categories, we cooperate with law enforcement on a case-by-case basis where preventing imminent harm is reasonably possible.

We will not, however, decrypt user vault contents we have no key for. As explained in our Privacy Policy, the architecture of the Service is designed so that we cannot read your stored conversations, files, or generated assets without your password. Legal process compelling us to produce data we cannot read will be answered with what we can produce (account metadata and ciphertext blobs) and a sworn explanation of why we cannot produce the plaintext.

14. Changes to This Policy

We may update this AUP from time to time to reflect new features, new abuse patterns, regulatory developments, or operational learnings. Updates are effective on the date posted at the top of this page. For material changes, we will give reasonable advance notice through the Service or by email to the address on file. Your continued use of the Service after an update takes effect constitutes acceptance of the updated AUP.

15. Contact

For questions about the AUP, contact us at:

• General trust and safety: [email protected]
• Abuse reports: [email protected]
• Account appeals: [email protected]
• Security disclosures: [email protected]

This Acceptable Use Policy is incorporated into the Terms of Service and works alongside the Privacy Policy.