Red Hills Operations provides AI employees designed to support business operations, improve customer experiences, and reduce repetitive work. We believe customers should understand when AI is being used, what information it may access, and where human oversight remains important.

Identification of AI Employees

Red Hills Operations is committed to transparent AI use. Customers should be informed when they are interacting with an AI-powered employee, assistant, or automated system where disclosure is appropriate.

Customer-Controlled Knowledge

Each business determines what information its AI employees are authorized to access and use. This may include business policies, approved documents, instructions, frequently asked questions, and connected business systems.

Customers remain responsible for reviewing the information, instructions, and permissions assigned to their AI employees.

In the owner-playground pilot, an eligible AI employee may retrieve bounded indexed excerpts only from active Knowledge collections assigned to that employee. Supporting citations identify customer-provided sources and persist with playground history. Historical citations may continue to identify an archived collection or an archived, superseded, or unavailable source version.

Approved AI Infrastructure Processing

When a customer enables searchable Knowledge Base or other AI capabilities, limited document excerpts, search queries, and related customer content may be securely processed by approved AI infrastructure providers. This may support document indexing, embedding generation, semantic search, AI employee knowledge retrieval, AI-generated responses, and other explicitly enabled AI capabilities.

Provider processing is limited to providing, operating, securing, and supporting those services under applicable contractual, technical, and privacy safeguards. Customer data is not sold, used for advertising, or intentionally shared with other customers. Using an approved provider does not authorize use of customer content as model-training data. API data is not used for provider model training by default unless a customer explicitly opts in through an approved process.

Application-controlled logs are designed to exclude document text, search queries, embeddings, and provider payloads. Approved providers may maintain separate technical telemetry under their own controlled terms; Red Hills Operations does not promise unverified zero-retention or Zero Data Retention treatment. A request setting that disables provider-side response storage does not constitute a universal zero-retention promise.

Private Business Information

Private customer information is used only to support the services requested by that customer. It is not used to benefit another customer, sold to third parties, or used to train public AI models.

Each organization's information is logically separated through tenant-based security controls.

Public Information and Secure Hybrid Intelligence

An AI employee may be permitted to use approved public information in addition to private business knowledge. Public information access is controlled through tool permissions and platform safeguards.

Private customer information must not be submitted to public web search tools. Public information access and approved AI infrastructure processing are separate controlled paths, each limited to the permissions and safeguards required for the enabled capability. Public-web search, external tools, and actions are not available in the owner-playground Knowledge pilot.

AI Limitations

AI-generated responses may occasionally be incomplete, inaccurate, or unsuitable for a particular situation. AI employees should not be treated as a substitute for qualified professional judgment in legal, medical, financial, safety-critical, or other high-risk matters.

Customers are responsible for deciding where human review or approval is required within their workflows.

Human Oversight

Red Hills Operations supports human oversight through configurable permissions, instructions, review processes, and escalation paths. Businesses should design workflows appropriate to the sensitivity and potential impact of each task.

Access and Tool Permissions

Human users, AI employees, integrations, and API credentials should receive only the minimum access necessary to perform their assigned work.

• AI employees may use only approved tools and information.
• Permissions may vary by employee and assigned responsibility.
• Customers may update or remove permissions as their needs change.
• Sensitive or higher-risk actions may require additional approval.

Monitoring and Auditability

Red Hills Operations may record platform activity such as logins, configuration changes, tool activity, API requests, and AI actions for security, troubleshooting, accountability, and service improvement.

Customer Responsibilities

• Provide accurate and lawful instructions to AI employees.
• Avoid uploading information the business is not authorized to use.
• Review permissions and connected systems regularly.
• Apply appropriate human review to consequential decisions.
• Notify Red Hills Operations of suspected misuse or security issues.

Updates to This Policy

We may update this AI Transparency policy as our services, capabilities, and safeguards evolve. Material changes will be reflected by updating the effective date on this page.

Contact Us

Questions regarding our use of artificial intelligence may be sent to:

support@redhillsoperations.com