The New York Times: A new OpenAI tool can recreate human voices — from just a 15-second recording. The technology, called Voice Engine, is currently available only to a small group of early testers. nyti.ms/3TGZIyS

A new OpenAI tool can recreate human voices — from just a 15-second recording. The technology, called Voice Engine, is currently available only to a small group of early testers. nyti.ms/3TGZIyS

OpenAI Unveils A.I. Technology That Recreates Human Voicesnyti.ms The start-up is sharing the technology, Voice Engine, with a small group of early testers as it tries to understand the potential dangers.

What could possibly go wrong?

Cool. Let’s weigh the pros and cons. Pros: NEAT, LOOK WHAT SCIENCE DID! Cons: Crime, wage theft, mass disinformation campaigns

Why.

RFK Jr., for one.

Because they want to push all creatives out of a job. I’m convinced.

Tech bros hate the fact that people don't think they're clever and funny, so they want to render all the people who are obsolete.

Might also be a backlash to unions popping up everywhere

Wouldn’t surprise me one bit. Although I’m in the UK who always had a slightly stronger union base than America but outside of railways, teaching and nurses I don’t know anyone in a union because they always roll over.

what could go right

You know, when Amish people said cameras would steal your soul they really were onto something there

What good purpose could such a technology have?

"So you can hear grandma again, without her consent."

I want all my voice messages to sound like Penelope Cruz.

uberduck.ai

Uberduck | Make Music with AI Vocalsuberduck.ai Generate high-quality voices by synthesizing your text with Uberduck's realistic voices or your own custom voices.

That ain't grandma, it's a computer skinwalking as her to emotionally manipulate people for tech corporations to profit off of. Creepy ass shit.

Well it's a godsend for people trying to run phone scams! Other than that, uh......

It was intended for people with speech disabilities, so text to speech could sound like their natural voice. It never occurred to me until now that someone would make a voice without the speaker’s consent.

Well, people are already scamming the boomers with copies of our voices saying we're in jail and need them to wire over x hundred dollars bail to these details right now. I read a thing by an actual lawyer who got scammed by that one, and that was before the voices got this good

Scam calls from your “family members” begging for money bc they’ve been “kidnapped/robbed and need money ASAP”

We are well beyond technology needing to have a good purpose. If it's possible we just do it and consequences be dammed

Wild that developing this tech doesn't come with a Conspiracy To Commit Fraud charge cuz wtf else would you use it for

Has it occurred to any of these people to make things that are actually useful for things other than scamming people or inaccurately cheating on homework?

It's not cheating if you still get the answer wrong.

It’s so nice they keep developing tools that can only turn things to shit. What a noble endeavor.

"In the future things will turn to shit" - Elon Musk

He’s certainly doing his part.

FUCKING WHY

So someone can scam a bunch of grandmas. There is no other use case…

Oh, I can see all kinds of use cases involving impersonating celebrity voices - unapproved political statements, unpaid endorsements, unapproved personal messages...

Thank god all my grandparents are dead.

But now with NanaAI, they don't have to be...

This also works for kidnapping scams for parents and guardians. And for bank scams.

I think this constantly.

Like my nanny could not manage this.

I've got an aunt who is the youngest of her siblings, who isn't. I'm not sure my dad, who is really tech savvy, will, tho he's been great with crypto and AI stuff overall. The George Carlin thing pissed him off. I'm not sure *I* am going to handle it.

Finally, a way for tech bros to once again hear the voices of the family members who are no longer speaking to them.

Awesome. I’m going to have Samuel Alito say a lot of very obscure slurs.

You could probably just put a directional microphone outside his dining room and get the same result.

If I knew where he lived I’m sure I could think of more creative things.

Another one for voice actors to band against? @caseymongillo.bsky.social

Finally, the Torment Nexus is here! I've been hoping for a long time that someone would get round to creating it. It seems really neat.

Awesome! The best uses for this tool are a) ripping off grandma and b) imitating celeb’s voices for advertising purposes. VERY EXCITING! 🤮

there is literally no good-faith use for the development of such tech.

I'm sure glad this company is working hard on a technology that, much like cryptocurrency, is almost exclusively useful for committing crimes.

Can’t wait to open my AI crammed phone with my AI generated icons to listen to my AI generated dead grandma begging me to pay her $50K in crypto for an AI written hospital bill

"The start-up is sharing the technology, Voice Engine, with a small group of early testers as it tries to understand the potential dangers." Tries to understand Tries Who are these people, that they don't know the dangers *immediately* Exactly what problem are we trying to solve here

Dumb. They're very very dumb.

Greedy also.

I hope they named this AI tool “Bababooey”.

Post