AI system overview

motel4.ai operates an AI-powered phone receptionist that answers inbound calls on behalf of lodging businesses. The system uses speech-to-text technology to transcribe caller speech in real time, a large language model to generate conversational responses based on property information provided by the business owner, and text-to-speech technology to deliver those responses audibly to the caller. The AI does not make autonomous decisions about reservations, pricing, or policies — it relays information configured by the property owner and captures guest inquiries for human follow-up.

AI identification and disclosure

Our AI receptionist identifies itself as an AI assistant at the beginning of every call. This proactive disclosure is built into the system and cannot be disabled. If a caller asks at any point whether they are speaking with a real person, the AI reconfirms that it is an artificial intelligence assistant and offers to connect the caller with a human. These practices satisfy disclosure obligations under the EU AI Act (Article 50), California AB 2905, Texas SB 140 (disclosure within 30 seconds), the Colorado AI Act, Illinois chatbot disclosure law, and pending FCC rules on AI-generated voice calls.

Data used by the AI

The AI generates responses using two categories of information.

• ✓Property information — name, address, room types, amenities, policies, check-in/check-out times, pricing, and local recommendations provided by the property owner through the dashboard
• ✓Conversation context — the caller's words transcribed in real time during the current call, used solely to generate relevant responses within that conversation

Data NOT used by the AI

The AI does not access previous callers' personal data to respond to new callers. Call transcripts and guest information from prior calls are not fed back into the AI model. We do not use any customer or guest data to train, fine-tune, or improve our AI models or any third-party AI models. The AI does not access external databases, credit reporting systems, or social media profiles.

AI limitations and accuracy

The AI receptionist generates responses probabilistically and may occasionally produce incorrect, incomplete, or inappropriate information. Specific limitations include:

• ✓The AI may misinterpret caller speech, especially in noisy environments, with strong accents, or with unusual proper nouns
• ✓The AI may provide outdated information if the property owner has not updated their property details in the dashboard
• ✓The AI cannot process payments, confirm reservations in external booking systems, or access real-time availability
• ✓The AI is not suitable for emergency situations — callers with emergencies should contact 911 directly
• ✓The AI may not correctly handle complex multi-part requests or negotiations

Human escalation

Any caller may request to speak with a human at any time during a call. When a caller requests human assistance, the AI will attempt to transfer the call to the fallback phone number configured by the property owner. If no fallback number is configured or the property owner is unavailable, the AI will take a message and flag it as urgent in the property owner's dashboard for prompt follow-up.

Automated decision-making

The AI receptionist does not make consequential decisions about individuals. It does not determine eligibility for services, set pricing based on caller characteristics, or make decisions that produce legal or similarly significant effects. All booking confirmations, pricing decisions, and service determinations are made by the property owner, not the AI. Under the California Automated Decision-Making Technology regulations (effective 2026), consumers have the right to opt out of automated decision-making that produces legal or similarly significant effects — because our AI does not make such decisions, this provision does not apply to standard use of the service.

Third-party AI providers

We use third-party AI service providers to deliver speech-to-text transcription, language model inference, and text-to-speech synthesis. These providers process call audio and text solely to generate responses in real time and are contractually prohibited from using this data for their own training or improvement purposes. For a full list of our service providers, see our Subprocessor List.

Your rights regarding AI processing

Depending on your jurisdiction, you may have specific rights regarding AI-processed data.

• ✓Right to human review — request that a human review any AI-generated summary or transcript for accuracy
• ✓Right to correction — request correction of inaccurate information in AI-generated transcripts
• ✓Right to deletion — request deletion of AI-generated transcripts and summaries
• ✓Right to explanation — request an explanation of how the AI generated a specific response
• ✓Right to opt out — if you are a guest and prefer not to interact with AI, request transfer to a human at the start of the call

Regulatory compliance

This notice is provided in compliance with the EU AI Act (Article 50, transparency obligations for AI systems interacting with natural persons), California AB 2905 (AI disclosure for automated phone systems), Texas SB 140 (AI voice disclosure within 30 seconds), the Colorado AI Act (consumer-facing AI deployer disclosure), Illinois chatbot disclosure requirements, and anticipated FCC rules on AI-generated voice communications. We monitor regulatory developments and update our practices and this notice accordingly. This notice was last updated June 2026.