AI Disclosure & Transparency

Last updated: April 11, 2026

How We Use Artificial Intelligence

Voxilo uses artificial intelligence (powered by xAI's Grok and Anthropic's Claude APIs) to conduct voice conversations with consumers who contact our clients' businesses. This page explains how our AI works, what it can and cannot do, and your rights.

AI Identification

Every first message in a conversation clearly discloses:

  • That you are communicating with an AI assistant, not a human
  • The name of the business on whose behalf the AI is acting
  • How to opt out (reply STOP)
  • How to request human follow-up (reply HUMAN or AGENT)

We will never misrepresent an AI-generated message as a human message. This disclosure complies with FCC proposed AI disclosure rules, CTIA messaging principles, and New York's Synthetic Performer Disclosure Law (effective June 2026).

What Our AI Does

  • Follows up on missed calls with qualifying questions
  • Gathers service details (problem description, location, urgency)
  • Schedules appointments based on client availability
  • Answers common FAQs using client-provided information
  • Detects emergencies and directs consumers to call 911

What Our AI Does NOT Do

  • Provide technical advice, diagnoses, or repair instructions
  • Quote prices or provide estimates (unless explicitly configured)
  • Make promises or commitments on behalf of the business
  • Process payments or access financial information
  • Provide medical, legal, or professional advice

AI Accuracy

AI-generated responses may contain inaccuracies. Our system includes quality scoring and drift detection to monitor AI performance, but no AI system is perfect. Important decisions should be confirmed directly with the business.

Your Rights

  • Request human follow-up at any time (reply HUMAN or AGENT)
  • Opt out of AI communication entirely (reply STOP)
  • Request your conversation transcript (email info@voxilo.ai)
  • Request deletion of your data (subject to legal retention requirements)

Data Processing

Conversation data is processed by xAI's Grok and Anthropic's Claude APIs in real time. Both providers' commercial API terms prohibit use of API inputs/outputs for training general-purpose models. Your conversations are not used to train AI. See our Privacy Policy for full data handling details.

Human Oversight

All AI interactions are subject to:

  • Automated quality scoring (5-dimension assessment)
  • Drift detection monitoring
  • Human review of flagged conversations
  • Weekly performance audits by the Voxilo operations team