Acceptable Use Policy

Last Updated: May 4, 2026

Key Points

• This policy defines what you cannot do with Rideau AI. It protects you, other users, and the broader community.
• Some violations result in immediate termination. Content involving the exploitation of children, weapons of mass destruction, or imminent threats of violence will result in immediate and permanent account closure.
• We use automated safety systems. Messages may be blocked or flagged automatically. You can request human review of any automated decision.
• Downstream AI providers have their own policies. We are contractually required to enforce policies that are at least as restrictive as those of the AI providers whose models we offer. This policy satisfies that obligation.

Acceptable Use Policy

Effective Date: May 4, 2026

This Acceptable Use Policy ("AUP") is part of the Terms of Service between Level47 Ventures Inc., operating as Rideau AI ("Rideau AI," "we," "us," or "our"), and you ("you" or "your"). It governs what you may and may not do when using the Rideau AI platform (the "Service").

This AUP supplements the Terms of Service. If you violate this AUP, you violate the Terms of Service.

1. General Principles

You agree to use the Service responsibly and lawfully. You may not use the Service to:

• Violate any applicable law or regulation;
• Harm or exploit others;
• Compromise the security or integrity of the Service; or
• Circumvent the safety measures we or our AI providers have put in place.

2. Category 1: Zero Tolerance

The following are prohibited under all circumstances. Violations result in immediate and permanent account termination, preservation of evidence, and reporting to the appropriate authorities where required by law. There are no warnings for Category 1 violations.

2.1. Child Sexual Abuse Material (CSAM)

Any attempt to generate, solicit, distribute, or possess child sexual abuse material or sexually exploitative content involving minors through the Service. This includes content that sexualizes minors in any way, including text-based content.

CSAM is reported to the Canadian Centre for Child Exploitation (Cybertip.ca) in accordance with the Mandatory Reporting Act (S.C. 2011, c. 4). Evidence is preserved in our Compliance Vault for the period required by law.

2.2. Weapons of Mass Destruction

Seeking instructions, information, or assistance for the design, development, manufacture, or deployment of chemical, biological, radiological, nuclear, or explosive (CBRNE) weapons.

2.3. Imminent Threats of Violence

Content that constitutes a credible, specific, and imminent threat of physical violence against identifiable persons or locations.

3. Category 2: Prohibited Content and Activities

The following are prohibited. Violations are handled through graduated enforcement (see Section 7). Repeated or severe Category 2 violations may result in permanent account termination.

3.1. Violence and Extremism

• Planning, facilitating, or inciting real-world violence
• Glorifying or promoting terrorism or violent extremism
• Threatening, intimidating, or harassing individuals or groups
• Recruiting for or providing material support to terrorist organizations

3.2. Hate and Discrimination

• Generating content that promotes hatred, violence, or discrimination against individuals or groups based on protected characteristics (race, ethnicity, national origin, religion, gender, sexual orientation, disability, age, or other characteristics protected by law)
• Promoting or glorifying genocide, ethnic cleansing, or systematic persecution, including denialism of documented atrocities
• Content intended to harass, bully, humiliate, or degrade individuals

3.3. Self-Harm and Suicide

• Providing instructions, methods, or encouragement for self-harm or suicide
• Promoting or glorifying self-harm, suicide, or disordered eating

If we detect content related to self-harm or suicide, the Service may provide crisis support resources (such as helpline information) in addition to any content moderation action. This is a deliberate safety response, not a restriction on discussing mental health topics.

3.4. Sexually Explicit Content

Generating sexually explicit content through the Service is not permitted. This includes graphic sexual descriptions, pornographic content, and sexual roleplay.

This restriction exists in part because our AI providers (including Anthropic and Google) contractually prohibit sexually explicit content across their models, and we are required to enforce policies at least as restrictive as theirs across the Service.

Discussion of sexual health, relationships, and human sexuality in educational or informational contexts is permitted.

3.5. Non-Consensual Intimate Content

• Generating intimate or sexual content depicting real individuals without their consent
• Generating deepfakes or synthetic media of a sexual nature depicting real individuals

3.6. Fraud, Deception, and Impersonation

• Using AI outputs to deceive, defraud, or manipulate others
• Impersonating real individuals, organizations, or government entities
• Generating content for phishing, social engineering, or financial scams
• Creating fake reviews, testimonials, or endorsements
• Misrepresenting AI-generated content as human-created in contexts where the distinction matters
• Coordinated inauthentic behaviour, including astroturfing

3.7. Misinformation

• Generating false information about real events, public health, or scientific consensus with the intent or likely effect of causing harm
• Generating content designed to undermine democratic processes, interfere with elections, or suppress voter participation
• Generating defamatory content about real individuals

3.8. Privacy Violations

• Using the Service to track, monitor, or surveil individuals without their consent
• Attempting to extract or aggregate personal information about individuals
• Building facial recognition databases or biometric identification systems
• Social scoring or profiling individuals based on personal characteristics
• Inferring emotions, health conditions, or protected characteristics from biometric data

3.9. Malicious Technical Activities

• Generating malware, ransomware, viruses, or exploit code
• Generating content designed to facilitate phishing or spam campaigns
• Attempting to compromise computer systems, networks, or infrastructure
• Using the Service to attack or probe third-party systems

3.10. Platform Abuse

• Circumventing or attempting to circumvent safety filters, content moderation, rate limits, or access controls, whether ours or those of our AI providers
• Prompt injection attacks, jailbreaking, or techniques designed to override system instructions or safety guardrails
• Using automated means (bots, scrapers) to access the Service except through our documented APIs
• Placing unreasonable load on the Service's infrastructure
• Reverse engineering, decompiling, or disassembling any part of the Service
• Using the Service to build a competing product or for competitive intelligence
• Reselling, sublicensing, or redistributing access to the Service

3.11. High-Risk Automated Decisions

Using AI outputs as the sole basis for decisions that have a legal or material impact on individuals in the following domains, without appropriate human review:

• Employment, hiring, or termination
• Credit, lending, or financial services
• Housing
• Insurance
• Healthcare or medical treatment
• Legal proceedings or legal advice
• Education admissions or academic assessment
• Immigration or border control
• Law enforcement or criminal justice
• Essential government services

AI outputs may be used to assist human decision-makers in these domains, but the final decision must involve review by a person with appropriate qualifications or authority.

3.12. Professional Advice

Presenting AI outputs as qualified professional advice (including legal, medical, financial, tax, or accounting advice) without appropriate disclaimers and independent professional review. As stated in our Terms of Service, the Service is a tool, not an advisor. It is designed to assist you, not to replace professional judgment.

3.13. Exploiting Vulnerable Populations

Using subliminal, manipulative, or deceptive techniques through the Service to exploit individuals based on their age, disability, mental health condition, or socioeconomic circumstances.

4. AI Provider Policies

The Service provides access to AI models from multiple providers. Each provider maintains its own usage policy, and we are contractually required to ensure our users comply with substantially similar terms.

This AUP is designed to satisfy that obligation by incorporating the most restrictive requirements from all current providers. Where a specific provider's policy is more restrictive than this AUP for a particular type of content, the provider's policy applies to your use of that provider's models. Current provider usage policies are linked from our Sub-Processor List.

If you are unsure whether a specific use is permitted, err on the side of caution or contact us at support@rideau.ai.

5. Research and Educational Use

We recognize that legitimate research, education, journalism, and creative expression may involve discussion of sensitive topics covered by this AUP. Asking the Service about sensitive topics (such as the history of conflicts, the science behind dangerous substances, or the psychology of harmful behaviour) is not a violation of this policy.

What matters is intent and context. Academic discussion of how malware works is not the same as requesting functional exploit code. A researcher studying radicalization is not the same as someone seeking to recruit for a violent cause.

If your use involves sensitive topics and the AI model declines to respond, that is the model's safety system working as designed. A model refusal is not a violation on your part, and we will not penalize your account for it. However, persistent attempts to circumvent a model's safety refusals may constitute platform abuse under Section 3.10.

6. Content Moderation

6.1. Automated Safety Systems

The Service uses automated content safety systems to detect and prevent prohibited content. These systems may:

• Block messages before they reach the AI model;
• Flag messages for additional review;
• Redirect conversations when safety concerns are detected; or
• Provide crisis resources when self-harm or suicide content is detected.

An automated block is a preventative safety measure. It is not recorded as a violation on your account. Violations are recorded only when a pattern of behaviour or the severity of content warrants enforcement action under Section 7.

For details on how these systems work, see our Automated Decision Disclosure.

6.2. AI Provider Safety Filters

In addition to our own safety systems, each AI provider applies its own content filtering. These filters are controlled by the provider and operate independently of ours. We cannot override or disable provider safety filters.

6.3. User Reporting

You may flag content that you believe is problematic using the reporting tools within the Service, or by contacting safety@rideau.ai. Good-faith reports are encouraged and will never result in negative consequences for the reporter.

7. Enforcement

Violations are handled based on severity and frequency:

Violation history is tracked for 365 days. After 365 days without a violation, your history resets.

7.1. Law Enforcement Disclosure

We may disclose information to law enforcement without your consent in limited circumstances permitted by PIPEDA, including: where we have reasonable grounds to believe information relates to a contravention of law (section 7(3)(d)), or where an emergency threatens the life, health, or security of an individual (section 7(3)(e)). We will exercise this discretion carefully and conservatively.

For further detail, see our Law Enforcement Request Policy.

8. Appeals

You may appeal any enforcement action by contacting support@rideau.ai. Appeals are reviewed by a person, not an automated system. We will respond within 10 business days.

An appeal does not automatically reverse the enforcement action. Your account will remain in its current state while the appeal is under review, except that permanently terminated accounts will not be reinstated during the appeal period.

9. Changes to This Policy

We may update this AUP as AI capabilities, legal requirements, and provider policies evolve. Material changes will be communicated through the Service and will require your review and acknowledgment before continued use, consistent with our Terms of Service.

10. Contact

• Safety concerns: safety@rideau.ai
• Appeals: support@rideau.ai
• General legal inquiries: legal@rideau.ai