Terms of Service (TOS)
1. Overview
cazimir.com/ provides an AI safety, monitoring, and alerting layer built on third-party automation platforms. The Service is provided “as-is” and “as-available.” By using the Service, you agree to all terms below.
2. Eligibility
You must be at least 18 years old and legally able to enter contracts.
3. Service Description
Cazimir monitors AI interactions and may generate alerts, logs, and risk assessments. Cazimir does not guarantee detection, prevention, suppression, or mitigation of harmful outputs or events. Cazimir is not real-time, not guaranteed accurate, and not guaranteed reliable.
4. No Warranty
Cazimir provides zero warranties of any kind. All services are used solely at your risk.
5. No Guarantee of Detection
Cazimir may fail to detect harmful content including, but not limited to, self-harm, violence, scams, medical misinformation, legal misinformation, extremism, bias, toxicity, and hallucinations.
6. Alerts & Notifications
Alerts may be delayed or never delivered. Cazimir is not responsible for SMS, email, webhook, or platform delivery failures.
7. Third-Party Dependencies
The Service relies on Voiceflow, Make.com, Twilio, Hume AI, Lakera/Guardrails, Airtable, Retool, and others. Cazimir is not responsible for outages, breaches, failures, inaccuracies, or issues originating from these platforms.
8. Data Handling
Data is routed through multiple third-party systems. You are responsible for obtaining user consent, managing storage, retention, deletion, and compliance.
User Responsibilities
- Maintain human oversight
- Implement your own crisis protocols
- Test your AI independently
- Configure alerts correctly
- Respond to incidents promptly
- Follow all laws and platform terms
10. Liability Limitation
To the fullest extent permitted by law:
Cazimir’s total liability is limited to the fees you paid in the past 30 days. Free or trial users are limited to $1 USD.
Cazimir is not liable for:
- Death, injury, self-harm, or suicide
- Emotional or psychological harm
- Financial loss, scams, or fraud
- Data breaches or leaks
- Regulatory violations
- Downtime or business interruption
- AI hallucinations or dangerous outputs
- Misuse by end-users
- Lost revenue or reputation
11. Indemnification
You agree to defend, indemnify, and hold Cazimir harmless from all claims arising from your AI outputs, end-user behavior, misuse, configuration choices, or regulatory violations.
12. Governing Law
This Agreement is governed by Wyoming law. All disputes must be resolved by binding arbitration in Wyoming. Class actions are prohibited.
13. Termination
Cazimir may suspend or terminate access at any time. All disclaimers and liability limits survive termination.
14. Entire Agreement
These terms supersede all prior agreements.
Privacy Policy
1. Data We Process
Cazimir processes:
- Messages between your AI and end-users
- Metadata (timestamps, events, technical details)
- Risk scores and behavioral analysis
- Error logs and alerts
2. How Data Is Processed
Data flows through third-party platforms including Voiceflow, Make.com, Hume AI, Lakera/Guardrails, Twilio, Airtable, and Retool. Cazimir does not control their infrastructure.
3. Data Storage
Cazimir does not permanently store messages unless your configuration in third-party services requires it. You control retention and deletion.
4. Lawful Basis & Consent
You must obtain consent from your users for message processing, AI analysis, and third-party routing.
5. Data Sharing
Data is shared only with integrated platforms and legal authorities where required.
6. Security Notice
Cazimir takes reasonable measures but does not guarantee zero breaches, zero leaks, uninterrupted service, or perfect security.
7. International Transfers
Your data may be processed globally depending on your chosen stack.
8. User Rights
You are responsible for handling end-user privacy requests. Cazimir does not interact with end-users directly.
9. Updates
This policy may be updated at any time. Continued use of the Service constitutes acceptance.
Disclaimers & Liability Waiver
No Duty of Care
Cazimir does not provide medical, legal, therapeutic, cybersecurity, or crisis-response services. Cazimir assumes no duty of care toward your users.
No Guarantee of Safety
AI safety cannot be guaranteed. All risk assessment features are approximate and may be incomplete or inaccurate.
High-Risk Use Cases
Any use of Cazimir in medical, mental-health, financial, legal, or life-critical contexts is strictly at your own risk.
Comprehensive Liability Waiver
By using Cazimir, you waive all claims against Cazimir for harm, injury, self-harm, suicide, financial losses, regulatory fines, data breaches, or any outcome generated by your AI system or your end-users.
Indemnification Notice
You agree to indemnify Cazimir Cazimir for all claims arising from your system, your users, or your practices.
End User License Agreement (EULA)
1. License Grant
Cazimir grants a non-exclusive, non-transferable license to use the Service for your business purposes.
2. Restrictions
- No reverse engineering
- No unauthorized resale
- No removal of notices or branding (unless approved)
- No illegal or prohibited use
3. Ownership
All code, workflows, models, and risk-scoring systems remain the property of cazimir.com/.
4. Updates
Updates may occur without notice. Cazimir is not obligated to maintain backward compatibility.
5. Termination
Violation of these terms terminates your license immediately.