The short answer
An AI employee speaks for your business on every call. Build in fairness, honesty and a clear path to a person from day one, so every caller gets the same respect and your team stays in control.
Why it matters
In May 2024, HUD issued Fair Housing Act guidance for tenant screening and housing advertising that use algorithms and AI (HUD). HUD has since withdrawn much of its fair housing guidance, including the advertising document, but the Fair Housing Act itself still applies to every answer (Fair housing and AI). NIST's voluntary AI Risk Management Framework, released in January 2023, gives a common structure for building trustworthy AI. Guardrails protect your residents and customers, and they protect your business.
The checklist
- Same answer for every caller. Pricing, availability, policies and next steps come from one approved source and never vary by who is asking.
- No guessing about people. The AI never infers or asks about protected characteristics such as race, religion, disability or family status, and never steers someone toward or away from a unit or offer.
- Say it's AI. Callers and chat users are told they're talking with an AI assistant, in plain words.
- A person is always one step away. Emergencies, complaints, accommodation requests and anything sensitive go straight to your team, with a clear summary.
- Stay in its lane. No legal, medical, financial or safety judgments. The AI captures the details and hands them to a qualified person.
- Consent and privacy first. Texts and follow-ups only go to people who opted in, personal details are stored securely, and recordings and transcripts follow your retention policy.
Keep it honest over time
- Review a sample of conversations every month. Look for answers that differ between callers, missed handoffs and confusing replies.
- Test with real scenarios. Ask the same question several ways, in different languages and with different names, and check the answers match.
- Write down what the AI does and doesn't do. Share it with your team so everyone knows when to step in.
- Update the knowledge base with every policy change. Outdated answers are a fairness problem as well as a service problem.
How YesOneOne helps
Our AI employees are trained on your approved policies and hand off to your team for anything sensitive, so the person on the other end gets a consistent, respectful answer and your people stay in control.