AI Terms & Conditions
1. Scope
These AI Terms supplement the Terms of Service ("General Terms") and govern the artificial-intelligence features of the mnnr platform: agent authorization and policy evaluation, AI-assisted analytics, anomaly and fraud-signal detection, the MNNR MCP server and agentic integrations, and any output generated with machine-learning or generative-AI systems ("AI Features"; such output, "AI Output"). By using any AI Feature you agree to these AI Terms. Where they conflict with the General Terms, these AI Terms control for the AI Features.
2. Description of AI Features
mnnr provides a rail-neutral control plane that authorizes and governs payments initiated by autonomous AI agents. AI Features may evaluate transactions against customer-defined policies, score risk and anomaly signals, summarize activity, and assist configuration. Some AI Features rely on third-party model providers (Section 9). mnnr does not itself hold, transmit, or settle funds; authorization decisions surfaced by AI Features are advisory signals executed under your configured policies.
3. Transparency — You Are Interacting With AI
Consistent with Article 50 of Regulation (EU) 2024/1689 (EU AI Act): certain interactions within the service are with an AI system, not a human; AI-generated content is identified as such where practicable; and where you deploy mnnr AI Features in your own products, you are responsible for the corresponding disclosures to your own end users where required by law.
4. Third-Party AI Agents and Assistants (ChatGPT, Gemini, Claude, Perplexity, and Others)
The service is designed to authorize and govern transactions initiated by AI agents, including agents built on or operated through third-party AI platforms such as OpenAI ChatGPT, Google Gemini, Anthropic Claude, Perplexity, and comparable systems ("Third-Party Agents"):
- Principal responsibility. Every Third-Party Agent acts on behalf of an identified account holder (the "Principal"). All actions, inputs, and transactions initiated by a Third-Party Agent are attributed to, and are the sole responsibility of, the Principal — exactly as if the Principal had acted directly.
- Agent authorization. Third-Party Agents may act only within the credentials, mandates, spend ceilings, and policies the Principal configured in mnnr. Agents acting outside mandate may be declined, throttled, or suspended without notice.
- Agent identification. Third-Party Agents must identify themselves as automated systems (accurate agent-identity headers; MCP or API credentials issued to the Principal). Impersonating a human or another principal's agent is prohibited.
- Platform terms. Use of a Third-Party Agent is also governed by that platform's own terms (OpenAI, Google, Anthropic, Perplexity, etc.). mnnr is not a party to those terms and is not responsible for any third-party AI platform's conduct, availability, or output.
- No agent-formed contracts beyond mandate. An instruction from your Third-Party Agent within its configured mandate is your instruction. mnnr is not liable for obligations an agent purports to create outside its configured mandate.
- Automated browsing. Third-Party Agents may read public pages for the Principal's use. Bulk scraping, model-training collection, and circumvention of rate limits or robots directives remain prohibited.
5. No Guarantee of Accuracy of AI Output
AI Output is generated probabilistically and is not guaranteed to be accurate, complete, current, or error-free. It may contain errors or hallucinations. Independently verify AI Output before relying on it; you are solely responsible for actions taken on its basis. AI Output is not legal, tax, accounting, investment, or compliance advice.
6. Human Oversight and Automated Decision-Making
Maintain meaningful human oversight appropriate to your use case. Where AI Features contribute to decisions with legal or similarly significant effects on natural persons (including payment declines), you are responsible for Article 22 GDPR and analogous laws — meaningful information about the logic involved, and rights to human intervention and contest. mnnr provides audit logs and policy-decision records to support these obligations.
7. Permitted and Prohibited Use
You must not, and must not permit any agent or third party to: engage in practices prohibited by Article 5 of the EU AI Act (manipulative, deceptive, or exploitative techniques; social scoring); use AI Features as the sole basis for high-risk decisions (credit, employment, housing, insurance, medical, legal) without required human review; submit inputs you lack rights to submit; use AI Features for fraud, money laundering, sanctions evasion, or unlawful payment activity; extract, reverse engineer, or distill the underlying models or use AI Output to train a competing model; circumvent safety mitigations or authorization controls; or misrepresent AI Output as human-generated where disclosure is required.
8. Input, AI Output, and Intellectual Property
You retain rights in your input. Subject to these AI Terms and third-party model terms, mnnr assigns to you its rights, if any, in AI Output generated for you. AI Output may not be copyright-protectable in some jurisdictions; similar output may be generated for other customers; mnnr retains all rights in the service, AI Features, and underlying systems.
9. Data Use, Training, and Third-Party Models
mnnr does not use your inputs or AI Output to train generalized foundation models. Aggregated, de-identified telemetry may be used to operate, secure, and improve the service. Certain AI Features are powered by third-party model providers under terms prohibiting training on your content (subprocessor list: privacy@mnnr.app). Personal data is processed per the Privacy Policy; EU deployments support EU data-residency controls.
10. Warranties Disclaimer
THE AI FEATURES AND AI OUTPUT ARE PROVIDED "AS IS" AND "AS AVAILABLE." TO THE MAXIMUM EXTENT PERMITTED BY LAW, MNNR DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, NON-INFRINGEMENT, ACCURACY, AND UNINTERRUPTED OPERATION.
11. Limitation of Liability
TO THE MAXIMUM EXTENT PERMITTED BY LAW, MNNR IS NOT LIABLE FOR LOSS ARISING FROM RELIANCE ON UNVERIFIED AI OUTPUT, FROM TRANSACTIONS EXECUTED BY YOUR AGENTS UNDER POLICIES YOU CONFIGURED, OR FROM ERRORS IN AI OUTPUT. AGGREGATE LIABILITY FOR AI FEATURES IS SUBJECT TO, AND COUNTS TOWARD, THE LIABILITY CAP IN THE GENERAL TERMS. NOTHING HERE EXCLUDES LIABILITY THAT CANNOT BE EXCLUDED BY LAW.
12. Indemnification
You will defend and indemnify mnnr against third-party claims arising from your inputs, your deployment of AI Features in your products, your failure to make legally required AI disclosures to your end users, or your violation of Sections 4 or 7.
13. Regulatory Change; Updates
AI law — including the EU AI Act and U.S. federal and state AI statutes — is evolving. mnnr may update these AI Terms with at least 30 days' notice of material changes; continued use constitutes acceptance. If you object, stop using the AI Features and, where a change materially degrades the service, terminate per the General Terms.
14. Suspension
mnnr may suspend or restrict AI Features immediately where reasonably necessary to prevent unlawful use, security risk, or violations of Sections 4 or 7, or where required by a regulator or model provider.
15. Governing Law
These AI Terms are governed by the same law and dispute-resolution provisions as the General Terms.