I created an interactive digital avatar of myself — and you can talk to it

A TechCrunch journalist has built and published an interactive AI avatar of herself, trained specifically to discuss venture fraud, and is now questioning whether personal AI clones are a good idea at all.

The Details

The experiment involved creating a lifelike digital replica capable of holding real conversations with anyone who visits it online. The avatar was trained on the journalist's voice, likeness, and knowledge base around a specific subject matter.

The result works. People can interact with it and get coherent, on-topic responses. But the creator came away with serious reservations rather than enthusiasm.

Key concerns raised include:

• Identity and consent boundaries: Who controls the avatar once it's live, and what happens when it says something the real person wouldn't?
• Misinformation risk: Even a well-intentioned clone can misrepresent nuanced positions or be taken out of context
• The "uncanny valley" of trust: Interacting with an AI that looks and sounds like a real person creates expectations the technology cannot reliably meet

The piece does not celebrate the technology. It treats the experience as a cautionary exploration more than a product endorsement.

Why This Matters for Service Providers

This experiment sits at the edge of where AI voice and avatar technology is heading, and it has direct implications for anyone deploying conversational AI on behalf of clients.

The core tension is trust and disclosure. Your SMB clients are already fielding questions from their own customers about whether they're talking to a real person. As AI becomes more human-sounding and human-looking, that question gets harder to answer honestly, and the regulatory and reputational stakes get higher.

For MSPs and telecom resellers building AI voice offerings, this is a reminder that QA and guardrails are not optional. A voice agent that says something wrong or out of character reflects on your client's brand, and by extension, on yours.

There is also a compliance dimension here. The line between "helpful AI assistant" and "deceptive impersonation" is being drawn right now through legislation and platform policy. Service providers who stay ahead of disclosure requirements will be better positioned when rules tighten, as they almost certainly will.

Looking Ahead

Watch for regulatory movement on AI persona disclosure requirements, particularly in sectors like healthcare, legal, and financial services where your clients already operate under scrutiny. If you are packaging AI voice agents for those verticals, build transparent disclosure language into the product now rather than retrofitting it later.

For the full story, read the original article on TechCrunch AI.