As we were walking past the Capitol building for the fourth time that day on November 14, in between meetings with lawmakers, I turned to our colleague Justin Toy from Platinum Advisors.

"Justin, do you want me to make a synthetic version of your voice?" I asked. 

"Sure," he said.

And without missing a beat, I pulled out my phone and began recording the conversation he was having with NAVA (National Association of Voice Actors) president, Tim Friedlander. I recorded about 10 seconds of Justin's voice and immediately uploaded it to Instant Voice Cloning on ElevenLabs.io. 

Within two minutes we had it. An almost perfect digital version of Justin's voice. 

We typed something silly about Platinum Advisors into the text box, and it spit out an incredibly believable sound file with breaths, exclamations, the right inflection - everything you would expect a human to sound like. 

So we sent it to Justin's boss. His boss shared it with the whole staff, and it took several minutes to inform everyone that it was a synthetic version of Justin. 

Thank God everyone appreciated the joke, because poor Justin might have been fired otherwise.

It now takes just three seconds of source audio to create a believable AI generated synthetic voice. Three seconds. Let that sink in. 

That means someone can record the outgoing voice message on your phone, create a synthetic version of you, and pretend to be you if they want to. They can make you say whatever they want.

This is not a "someday" scenario. This is possible RIGHT NOW.

As of this writing, your voice is not federally protected in the United States. And there is no federal law that says you own the rights to your voice. You DON'T own the rights to your own voice. This is why we at NAVA are fighting.

Huge decisions are being made around AI right now. Legislation is being written, contracts are being written, regulations are being written, court cases are pending, and we want and NEED our elected officials to take us into consideration.

NAVA went to Washington DC on November 13 and 14  to meet with 12 different offices of senators and representatives over two days. The lawmakers we met with are deeply interested in technology and AI, and seem to want regulation as much as we do.

It was so encouraging to sit with their staff while they took notes and really listened to what we had to say. 

The first thing that comes out regulating AI will set the precedent for ALL OTHER REGULATIONS. That is true for federal legislation, state legislation, court cases, and widely adopted contracts. Our lawmakers want to hear what we want. They want to know what we are okay with.

Consent. Control. Compensation.

These are the three things that need to be included EXPLICITLY in any law, any court case, and any contract.

It is really not that difficult to get consent. There are thousands of people in the world who will willingly train synthetic voice models for a fee. Some would even do it for free.

But there are many, many people who don't want their voices to be used in that way.

And by using their sound files without their explicit consent, developers are robbing them of the ability to choose. 

Control. I want to know if my voice is being used in a Trump ad. Or an ad for Pepsi. Or pornography. Or as the new voice of TikTok. 

I want the company that made my synthetic voice to tell me if they are being bought out by another company, or if they intend to sell my voice to another company. I want control over how my voice is used. 

Some people don't. They probably don't make a living with their voices. But they still need the choice. 

And compensation. This one is a little trickier than the first two. There isn't a standard method of compensation for synthetic voice clones yet. But the standard will likely be based on number of impressions, or per unit generated, or even a yearly licensing fee. Or even a buyout. 

NAVA is working with actors and AI companies to develop standard methods of compensation. But for the time being let's just agree that fair compensation means both sides of the equation feel that they are getting a fair deal. 

What we really need are laws. Laws that protect our likeness and image. Laws that give the right of publicity to all people across the nation and around the world. And they are coming. 

NAVA is tackling one side of the fight by meeting with the people writing the laws. But until then it's up to us... on both sides... to protect each other. Because if someone wants to use my voice... it should be my choice. 

That's why NAVA went to Washington. And will be back as many times as it takes.