AI Usage & Responsible AI Policy

Purpose

ControlShift AI ("ControlShift AI," "we," "our," or "us") builds AI-powered Chat Agent and Voice Agent platforms. This policy describes how we develop, deploy, and govern artificial intelligence in our products and services, and the responsibilities of customers who use them.

Our Responsible AI Principles

• Transparency: We aim to make clear when users are interacting with an AI system and what data may be processed.
• Human oversight: Customers retain control over agent behavior, knowledge sources, and escalation paths to human staff.
• Privacy and security: AI processing is subject to our Privacy Policy, Security Policy, and applicable data protection laws.
• Fairness and non-discrimination: We work to reduce harmful bias in system design and encourage customers to monitor outputs in sensitive contexts.
• Accountability: We maintain internal review processes for AI features and respond to reported misuse or harmful outputs.

How We Use AI

Our platform may use AI technologies to:

• Generate conversational responses in chat and voice agents
• Summarize, classify, or route customer inquiries
• Assist with workflow automation and integrations
• Improve product performance through aggregated usage insights

AI outputs are probabilistic and may be inaccurate or incomplete. Customers are responsible for reviewing AI-generated content before relying on it for legal, medical, financial, or other high-risk decisions.

Customer Responsibilities

When using ControlShift AI, you agree to:

• Configure agents appropriately for your industry and use case
• Disclose AI usage to end users where required by law or regulation
• Avoid deploying agents for prohibited uses described in our Acceptable Use Policy
• Obtain valid consent before recording voice interactions where required (see our Voice Recording Consent Policy)
• Monitor agent behavior and promptly report safety or compliance concerns

Prohibited AI Uses

You must not use our AI services to:

• Generate deceptive, fraudulent, or impersonation content
• Produce unlawful, harassing, hateful, or exploitative material
• Make fully automated decisions with legal or similarly significant effects without appropriate safeguards
• Process special category data without a lawful basis and appropriate controls
• Circumvent safety filters, usage limits, or platform restrictions

Data Used for AI Processing

AI features may process prompts, conversation transcripts, knowledge base content, and integration data provided by customers. We do not use customer content to train public foundation models unless explicitly agreed in writing. See our Data Processing Agreement for processor obligations where applicable.

Model Providers and Subprocessors

We may rely on third-party AI infrastructure and model providers to deliver our services. A current list of subprocessors is available on our Subprocessor List.

Incident Reporting

If you encounter harmful, biased, or unsafe AI behavior on our platform, contact us at safety@controlshiftai.com.au or support@controlshiftai.com with relevant details. We will investigate and take appropriate action.

Policy Updates

We may update this policy as AI capabilities, regulations, and best practices evolve. Material changes will be posted on this page with an updated "Last updated" date.

Contact Us

• Email: support@controlshiftai.com
• Safety: safety@controlshiftai.com.au
• Address: Sydney, NSW, Australia