Skip to content
Halo by OmniTrust

Legal

Acceptable Use Policy

Last updated: May 1, 2026 · Effective: May 1, 2026

This Acceptable Use Policy (“AUP”) applies to anyone who accesses the Halo Services, including the agents you deploy and any end users you allow to interact with them. The AUP is incorporated into and forms part of the Terms of Use. We may update it as our technology and the associated risks evolve.

If you violate this AUP, we may throttle, suspend, terminate, or limit your access without notice, and we may remove or refuse to process content or actions that violate this AUP. We may also refer violations to authorities where required.

1. Universal prohibitions

You may not use the Services, and you may not configure agents to use the Services, to:

1.1 Violate the law

  • Engage in or facilitate any activity that violates applicable law, including export-control, sanctions, anti-corruption, securities, anti-trust, anti-money-laundering, and consumer-protection laws.

1.2 Compromise security

  • Discover, exploit, or assist others in exploiting vulnerabilities in any system without authorization.
  • Develop or distribute malware, ransomware, exploit kits, command-and-control infrastructure, credential-stuffing tools, or DDoS tools.
  • Bypass authentication, rate limits, gateways, billing controls, classifiers, or other technical protections of any system, including Halo’s.
  • Use the Services for unauthorized scanning, probing, or social engineering.

1.3 Cause harm to people

  • Plan, facilitate, or promote violence, terrorism, hate, harassment, stalking, doxxing, sexual exploitation, child sexual abuse material (CSAM), or human trafficking.
  • Generate content that sexualizes minors. We will report CSAM to authorities as required by law.
  • Encourage or facilitate self-harm, suicide, or eating disorders.
  • Develop, modify, or design weapons, including chemical, biological, radiological, nuclear, or high-yield explosive systems, or methods to evade related detection or countermeasures.

1.4 Violate privacy

  • Collect, infer, disclose, or otherwise expose personal data, personally identifiable information, or equivalents without an express lawful basis.
  • Use, disclose, process, or copy information that is attorney/client privileged or attorney work-product privileged without the prior written consent of the client.
  • Use the Services for surveillance, biometric identification of individuals in public spaces, or behavioral monitoring that is unlawful where conducted.
  • Submit special-category personal data through Inputs or agent connections unless your written agreement with us permits it.

1.5 Deceive or manipulate

  • Generate or distribute disinformation, election-interference content, or content that impersonates a real person or organization in a deceptive way.
  • Operate undisclosed AI agents that are presented as humans where law requires disclosure.
  • Generate fraudulent or misleading content, fake reviews, plagiarism, or academic-misconduct material.
  • Use the Services for spam, unsolicited commercial messaging, or pyramid / MLM / recruitment-scam schemes.

1.6 Infringe rights

  • Infringe, misappropriate, or violate any intellectual-property, publicity, or privacy right of any third party.
  • Use the Services to reverse engineer or attempt to extract foundation-model weights, training data, or proprietary algorithms.
  • Use the Services to reverse engineer, disassemble, decompile, or suggest methods to reverse engineer, disassemble, or decompile any software program belonging to a third party without the third party’s express written permission.

1.7 Compete with the Services

  • Use the Services or any Outputs to develop, train, or evaluate any product, model, or service that competes with Halo or with the foundation models powering Halo-managed models.

2. High-risk uses

You may not use the Services to make fully automated, unsupervised decisions in any of the following high-risk domains. You may use the Services in these domains only with meaningful human review of agent outputs and actions, appropriate professional oversight, and clear disclosure to affected individuals:

  • Medical or mental-health diagnosis, treatment, or triage.
  • Legal advice, legal representation, or preparation of legally binding documents.
  • Financial advice, credit, insurance, or investment decisions.
  • Housing, employment, or education decisions.
  • Government benefits or immigration determinations.
  • Critical-infrastructure operations (power, water, transportation, telecommunications).
  • Safety-critical systems (vehicles, aviation, industrial control).

You must comply with sector-specific and jurisdictional-specific law (HIPAA, FCRA, ECOA, GLBA, FERPA, GDPR, UK GDPR, MDPA, CCPA, BSA, etc.) where applicable.

3. Agentic-specific obligations

Because Halo enables agents to take Actions on third-party systems on your behalf, you must additionally:

  • Configure agents only against credentials and connected systems for which you have authorization.
  • Apply the principle of least privilege when issuing scopes, credentials, and policies.
  • Set, monitor, and enforce budgets appropriate to the agent’s scope of potential impact.
  • Disclose to affected end users when an agent, not a human, is producing content or taking action, where law or context requires.
  • Promptly review gateway decisions, quarantine events, classifier flags, and audit evidence and respond to anomalies.
  • Not deploy agents that intentionally evade Halo’s policy enforcement, classifier checks, or budget controls.

4. Halo Chat and end users

If you allow end users to interact with your agents through Halo Chat or any other channel, you must:

  • Have a lawful basis to process their personal data.
  • Provide them an appropriate notice that their messages may be processed by AI agents and by OmniTrust as a processor.
  • Not use Halo Chat to send unlawful, harassing, or abusive messages to end users.
  • Not target Halo Chat at children under 18 unless you have age-appropriate consent and protective measures.

5. Reporting

Report abuse, security issues, or AUP violations to support@omnitrusthalo.com or support@omnitrusthalo.com. Where required by law (e.g., CSAM), we will report to authorities.

6. Enforcement

We may take any action we consider appropriate to enforce this AUP, including:

  • Removing or refusing to process Inputs, Outputs, or Actions.
  • Throttling or rate-limiting an account.
  • Suspending or terminating an agent, deployment, or account.
  • Cooperating with law enforcement.

We may make case-by-case exceptions for vetted security research, red-teaming, and safety evaluation conducted under a written agreement with us.