Building AI that calls or texts customers? Start with the legal workflow, not just the demo.
I’m John Henson, founder of Henson Legal PLLC. I advise businesses on TCPA, AI voice, and lead-gen compliance.
I talk about that (and other stuff) here.
https://t.co/TChwezWBYW
@BrianKeith@H2Hyperbaric@AJA_Cortes@bot@grok Considering I have about 5 clients doing a variant of this (in healthcare and other sectors), then yes. It is worth it for some people.
@andrixx_code Smart. People are thinking of compliance as an afterthought, particularly with AI voice. The risk is too high and the fixes are too simple.
@LingoAITech Part of this is already settled. Under UETA §14, a contract formed by your 'electronic agent' can bind you even if no human reviewed it. That's a matter for terms of service and the application of agency law. That's where the fights will be.
@FynCas "Looks 100% human" is exactly the line to watch for some state laws. NY requires a clear disclosure when an ad knowingly uses a synthetic performer, CA's version starts Jan. 1, 2027. In NY could be up to $5k per ad penalty
@voicexprt Two US details worth adding: an AI voice message has to say who's calling at the start, and once the call sells anything, you need prior express written consent, not just 'they agreed to reminders.' Keep reminder agents strictly informational.
@BrianRoemmele Nice build. The fact you are disclosing recorded line is huge, because a lot of people miss that. Also, make sure to avoid ANY marketing language or you could have TCPA issues.
@seancassy@DealershipAttny I deal with this issue every day. Old service consent to text the customer is NOT consent to use AI voice. They are two different things. This is a problem clients are having. Old consent models don't necessarily fit the current needs of the business.
@pavelsukhachev I'd add consent before the dial, not just disclosure after pickup. An AI voice making a sales call needs prior express written consent, and insurance has its own state telemarketing rules on top.