Artificial Intelligence Acceptable Use Policy

Last Modified: October 22, 2025

This Artificial Intelligence Acceptable Use Policy (“Policy”) governs the use of Iru’s AI-powered products, including but not limited to Kai. This Policy applies to all Iru customers, their employees, and any third-party partners interacting with AI functionalities within Iru’s services. The Policy supplements the Master Subscription Agreement.

This Policy was last updated on October 22, 2025.

Iru reserves the right to modify this Policy at any time by posting the updated version on https://www.iru.com/legal/ai-use-policy/. Changes will be effective upon posting. Any violation of this Policy may be considered a breach of the Master Subscription Agreement or other governing agreements.

AI-related features within Iru’s Services must not be used for the following purposes:

AI-driven decisions must not be made without human oversight when such decisions have legal or significantly impactful consequences. In cases where Iru AI functionalities provide recommendations (e.g., computer fleet configurations affecting compliance), a human must make the final decision, considering other factors.

6. Individualized Advice from Licensed Professionals

Iru AI functionalities must not be used to generate or provide personalized advice that would typically be offered by licensed professionals, such as legal, financial, or medical experts. If Iru AI functionalities assist in providing non-individualized information, this must be explicitly stated, and a qualified human must review the AI output if used in decision-making.

7. Sensitive Data Prediction or Profiling

Iru AI functionalities must not be used to predict or infer any sensitive personal information about users, including but not limited to race, ethnicity, political beliefs, religion, gender, or health conditions. Additionally, it may not process biometric data (such as fingerprints or facial geometry) for profiling or identification purposes unless explicitly allowed by the customer under applicable laws.

8. Deceptive or Misleading Activities

Iru AI functionalities may not be used to engage in activities aimed at manipulating or misleading users, such as distributing false or deceptive content, including deepfakes or highly edited media misrepresented as genuine. Iru AI generated content must be clearly identified as AI-generated when interacting with end users unless obvious from context.

9. Child Exploitation and Abuse

Iru AI functionalities must not be utilized for any activities related to child exploitation, including generating, distributing, or enabling access to Child Sexual Abuse Material (CSAM).

10. Weaponization

Iru AI functionalities may not be used for the development, distribution, or sale of weapons, explosives, or other military-grade technologies.

11. Adult Content and Dating Apps

Iru AI functionalities may not be used to generate, distribute, or assist with sexually explicit content, non-consensual intimate imagery, deepfake pornography, or adult chatbots.

12. High-Risk Use Disclosure

Customers must assess whether using Iru’s AI functionalities for high-risk applications is appropriate, especially in contexts that could result in severe outcomes (e.g., physical harm or system failures in critical infrastructure). Iru’s AI products are not designed for high-risk scenarios where failure could result in significant damage, injury, or death.

13.  Transparency and Disclosure

Iru AI interactions with end users must be transparent. Iru must disclose when an end user is interacting with Iru AI functionalities unless it is evident from the context (e.g., chatbot interactions). Iru encourages users to implement human-in-the-loop systems when using Iru AI functionalities for decision-making in sensitive or high-stakes scenarios.

This policy ensures ethical, transparent, and safe use of AI within Iru AI functionalities and aligns with industry standards for responsible AI deployment. For questions or concerns about this Policy or Iru AI functionalities, please contact legal@iru.com.