- OpenAI approached Scarlett last fall, and she refused.
- Two days before the GPT-4o launch, they contacted her agent and asked that she reconsider. (Two days! This means they already had everything they needed to ship the product with Scarlett’s cloned voice.)
- Not receiving a response, OpenAI demos the product anyway, with Sam tweeting “her” in reference to Scarlett’s film.
- When Scarlett’s counsel asked for an explanation of how the “Sky” voice was created, OpenAI yanked the voice from their product line.
Perhaps Sam’s next tweet should read “red-handed”.
To me it's about as close to her voice as saying "It's a woman's voice". Not to say all women sound alike but the sound I heard from that video above could maybe best be described and "generic peppy female American spokesperson voice"
Even listening to it now with the suggestion that it might sound like her I don't personally hear Scarlett Johansson's voice from the demo.
There may be some damming proof where they find they sampled her specifically but saying they negotiated and didn't come to an agreement is not proof that it's supposed to be her voice. Again, to me it just sounds like a generic voice. I've used the the version before GPT-4o and I never got the vibe it was Scarlett Johansson.
I did get the "Her" vibe but only because I was talking to a computer with a female voice and it was easy to imagine that something like "Her" was in the near future. I also imagined or wished that it was Majel Barrett from ST:TNG, if only because the computer on ST:TNG gave short and useful answers where as ChatGPT always gives long-winded repetitive annoying answers
"Midler was asked to sing a famous song of hers for the commercial and refused. Subsequently, the company hired a voice-impersonator of Midler and carried on with using the song for the commercial, since it had been approved by the copyright-holder. Midler's image and likeness were not used in the commercial but many claimed the voice used sounded impeccably like Midler's."
As a casual mostly observer of AI, even I was aware of this precedent
New voice mode is a speech predicting transformer. "Voice Cloning" could be as simple as appending a sample of the voice to the context and instructing it to imitate it.
OpenAI first demoed and launched the “Sky” voice in November last year. The new demo doesn’t appear to have a new voice.
I doubt it would take them long to prepare a new voice, and who’s to say they wouldn’t delay the announcements for a ScarJo voice?
A charitable interpretation of the “her” tweet would be a comparison to the conversational and AI capabilities of the product, not the voice specifically, but it’s certainly not a good look.
…Unless, of course, all this scandal isn't also a part of marketing campaign.
Apparently they had no confidence in defending themselves, so why even release with the voice in the first place?
Seems pretty reckless to not have alternatives just in case Scarlett refused.
Still live for me? Unless the Sky I’m getting is a different one?
Don't hear any arguments on how this is fair-use. (It isn't)
Why? Because everyone (including OpenAI) knows it clearly isn't fair-use even after pulling the voice.
Sam should be ashamed to have ever thought of ripping off anyone's voice, let alone done it and rolled it out.
They are building some potentially world-changing technology, but cannot rise above being basically creepy rip-off artists. Einstein was right about requiring improved ethics to meet new challenges, and also that we are not meeting that requirement.
sad to see
The movie industry does this all the time.
Johansson is probably suing them so they're forced to remove the Sky voice while the lawsuit is happening.
I'm not a fan of Sam Altman or OpenAI but they didn't do anything wrong here.
But I’ve defended them from unfair criticism on more than a few occasions and I feel that of all the things to land on them about this one is a fairly mundane screwup that could be a scrappy PM pushing their mandate that got corrected quickly.
The leadership for the most part scares the shit out of me, and clearly a house-cleaning is in order.
But of all the things to take them to task over? There’s legitimately damning shit this week, this feels like someone exceeded their mandate from the mid-level and legal walked it back.
For example, a car company approached the band sigur ros to include some of their music in a car commercial. Sigur ros declined. A few months later the commercial airs with a song that sounds like an unreleased sigur ros song, but really they just paid a composer to make something that sounds like sigur ros, but isn't. So maybe openai just had a random lady with a voice similar to Scarlett so the recording.
Taking down the voice could just be concern for bad press, or trying to avoid lawsuits regardless of whether you think you are in the right or not. Per this* CNN article:
> Johansson said she hired legal counsel, and said OpenAI “reluctantly agreed” to take down the “Sky” voice after her counsel sent Altman two letters.
So, Johansson's lawyers probably said something like "I'll sue your pants off if you don't take it down". And then they took it down. You can't use that as evidence that they are guilty. It could just as easily be the case that they didn't want to go to court over this even if they thought they were legally above board.
* https://www.cnn.com/2024/05/20/tech/openai-pausing-flirty-ch...
> Though the most successful founders are usually good people, they tend to have a piratical gleam in their eye. They're not Goody Two-Shoes type good. Morally, they care about getting the big questions right, but not about observing proprieties. That's why I'd use the word naughty rather than evil. They delight in breaking rules, but not rules that matter. This quality may be redundant though; it may be implied by imagination.
> Sam Altman of Loopt is one of the most successful alumni, so we asked him what question we could put on the Y Combinator application that would help us discover more people like him. He said to ask about a time when they'd hacked something to their advantage—hacked in the sense of beating the system, not breaking into computers. It has become one of the questions we pay most attention to when judging applications.
"What We Look for in Founders", PG
https://paulgraham.com/founders.html
I think the more powerful you become, the less endearing this trait is.
I mention this specifically because I remember mark andreseen comment something similar in lex fridman's podcast, something along the lines of getting "those creative people" together to build on ai.
> They delight in breaking rules, but not rules that matter.
The question becomes "what rules matter?". And the answer inevitably becomes "only the ones that work in my favor and/or that I agree with".
I think someone trying to defend this would go "oh come on, does it really matter if a rich actress gets slightly richer?" And no, honestly, it doesn't matter that much. Not to me, anyway. But it matters that it establishes (or rather, confirms and reinforces) a culture of disregard and makes it about what you think matters, and not about what someone else might think matters about the things in their life. Their life belongs to them, a fact that utopians have forgotten again and again everywhere and everywhen. And once all judgement is up to you, if you're a sufficiently ambitious and motivated reasoner (and the kind of person we're talking about here is), you can justify pretty much whatever you want without that pesky real-world check of a person going "um actually no I don't want you to do that".
Sometimes I think anti-tech takes get this wrong. They see the problem as breaking the rules at all, as disrupting the status quo at all, as taking any action that might reasonably be foreseen to cause harm. But you do really have to do that if you want to make something good sometimes. You can't foresee every consequence of your actions - I doubt, for example, that Airbnb's founders were thinking about issues with housing policy when they started their company. But what differentiates behavior like this from risk-taking is that the harm here is deliberate and considered. Mistakes happen, but this was not a mistake. It was a choice to say "this is mine now".
That isn't a high bar to clear. And I think we can demand that tech leaders clear it without stifling the innovation that is tech at its best.
To them*
Which is the whole problem. These narcissistic egotists think they, alone, individually, are capable of deciding what's best not just for their companies but for humanity writ large.
If so, I suspect they’ll be okay in a court of law — having a voice similar to a celebrity isn’t illegal.
It’ll likely cheese off actors and performers though.
[1] https://www.forbes.com/sites/roberthart/2024/05/20/openai-sa...
Of course, Twitter continues to bring people with big egos to their own downfall.
But it must feel pretty fucking weird and violatory when you spend your entire life thinking about how you are going to deliver certain lines and that’s your creative Body of work, and then for someone to just take that Voice and apply it to any random text that can be generated?
I get why she wouldn’t want to let it go.
In a way it is similar to how a developer might feel about their code being absorbed, generalized, and then regurgitated almost verbatim as part of some AI responses
But in the case of voice it’s even worse as the personality impression is contained in the slightest utterance… whereas a style of coding Or a piece of code might be less Recognizable, and generally applicable to such a wide range of productions
Voice is the original human technology, To try to take that from someone without their consent is a pretty all encompassing grab
Not a bad call for someone already rich
And in terms of collaboration potential... OpenAI is a big draw for businesses and a subset of tech enthusiasts, but I don't think artists in any industry are dying to collaborate with them.
We know nothing about their offer to her. Could have just been a bad deal
They changed the voice to intone like Scarlett Johansson's character. It's like they changed the song the voice was singing to one that lots of people recognise.
https://news.ycombinator.com/item?id=40414908
https://news.ycombinator.com/item?id=40414923
https://news.ycombinator.com/item?id=40419791
https://news.ycombinator.com/item?id=40414802
https://news.ycombinator.com/item?id=40414902
https://news.ycombinator.com/item?id=40414713
https://news.ycombinator.com/item?id=40415350
Saw some other posters expressing this view who deleted their posts after getting downvoted, lol.
Stuff from her comes via press agents, which is generally sent directly to reporters.
[1] https://variety.com/2024/digital/news/openai-pulls-scarlett-...
One thing these trained voices make clear is that it's a tts engine generating ChatGPT-4o's speech, same as before. The whole omni-modal spin suggesting that the model is natively consuming and generating speech appears to be bunk.
I'm not familiar with the specifics of how AI models work but doesn't the ability from some of the demos rule out what you've said above? Eg. The speeding up and slowing down speech and the sarcasm don't seem possible if TTS was a separate component
Nevertheless. This is still incredibly embarrassing for OpenAI. And totally hurts the company’s aspiration to be good for humanity.
This doesn't make any sense. If it's a speech to speech transformer then 'training' could just be a sample at the beginning of the context window. Or it could one of several voices used for the Instruct-tuning or RLHF process. Either way, it doesn't debunk anything.
He's using trade secrets, copyright, patents, NDAs liberally.
This is not a principled stand, just opportunism.
They have no moat, they can't fix hallucinations, and people are starting to realize it's nowhere near as useful or close to AGI as he's been saying. If they hate him too, this ship is sunk.
What a bloody arrogant idiot.
The wink wink at creating an AI girlfriend is so bizarre
I guess we know who their target user base is
https://twitter.com/OpenAI/status/1790089521985466587
Giggly, flirty AI voice demos were already weird, but now it's even creepier knowing the backstory of how they try to get their voices.
So he's here to help regulate it all with an "international agency" (see the reference[2] by windexh8er in this thread)! Don't forget that Altman is the same hack who came up with "Worldcoin" and the so-called "Orb" that'll scan your eyeballs for "proof of personhood".
Is this sleazy marketer the one to be trusted to lead an effort that has a lasting impact on humanity? Hell no.
It’s so refreshing to hear someone else actually understand this sentiment for what it is— snake oil sales. The normies out there eat this up, and there is no convincing them otherwise because of how powerful the AI trope in entertainment media is.
Honestly though, if you actually listen to him and read his words he seems to be even more devoid of basic empathetic human traits than even Zuckerberg who gets widely lampooned as a robot or a lizard.
He is a grifter through-and-through.
And hasn't OpenAI recently shown that they can pull off a commercial coup d'état, unscathed?
Why would they not simply also take the voice of some actress? That's small potatoes.
No one is going to push back against OpenAI meaningfully.
People are still going to use ChatGPT to cheat on their homework, to phone-in their jobs, and to try to ride OpenAI's coattails.
The current staff have already shown they're aligned with the coup.
Politicians and business leaders befriend money.
Maybe OpenAI will eventually settle with the actress, for a handful of coins they found in the cushions of their trillion-dollar sofa.
It sure was. But OpenAI decided to poke the Bear and is being sued by NYT. And apparently as a sidequest they thought it best to put their head in a lion's mouth. I wouldn't call the PR clout and finances of an A-list celebrity small potators.
They could have easily flown under the radar and have been praised as the next Google if they kept to petty thievery on the internet instead of going for the high profile content.
>People are still going to use ChatGPT to cheat on their homework, to phone-in their jobs, and to try to ride OpenAI's coattails.
Sure, and ChatGPT isn't goint to make lots of money from these small time users. They want to target corporate, and nothing scares of coporate more than pending litigation. So I think this will bite them sooner rathter than later.
>Maybe OpenAI will eventually settle with the actress, for a handful of coins they found in the cushions of their trillion-dollar sofa.
I suppose we'll see. I'm sure she was offered a few pennies as is, and she rejected that. She may not be in it for the money. She very likely doesn't need to work another day in her life as is.
Couldn't, perhaps, one of the more famous people on Earth be responsible for "meaningfully" taking OpenAI to task for this? Perhaps even being the impetus for legislative action?
Altman and OpenAI will walk over everyone here without any difficulty if they decide to take whats ours.
Why would people not want laws? The answer is so they can do the things that the laws prevent.
This is POSIWID territory [0]. "The purpose of a system is what it does". Not what it repeatedly fails to live up to.
What was the primary investment purpose of Uber? Not any of the things it will forever fail to turn a profit at. It was to destroy regulations preventing companies like Uber doing what they do. That is what it succeeded at.
The purpose of OpenAI is to minimise and denigrate the idea of individual human contributions.
[0] https://en.wikipedia.org/wiki/The_purpose_of_a_system_is_wha...
She has the resources to fight back and make an example of them, and they have the resources to make it worthwhile.
The fact that they reached out to her multiple times and insinuated it was supposed to sound like her with Sam's "her" tweet makes a pretty clear connection to her. Without that they'd probably be fine.
Bette Midler sued Ford under very similar circumstances and won.
Scarlett Johansson's character in the movie is not Scarlett Johansson although her voice is very similar. I wouldn't say it's identical.
It is quite possible that OpenAI has synthesized the voice from SJ material.
However If OpenAI can produce the woman who did is the current voice, and she has a voice nearly identical that of SJ would that mean OpenAI had done something wrong?
Does SJ since she is a celebrity hold a "patent" right to sound like her.
The more likely scenario is that they have hired a person and told her to try and imitate how SJ sounds.
What is the law on something like that?
And would they need to use a voice actor when there is a substantial body of movie dialogue and interviews? I'd be surprised if they'd bothered.
> But OpenAI's chief technology officer, Mira Murati, has said that GPT-4o's voice modes were less inspired by Her than by studying the "really natural, rich, and interactive" aspects of human conversation, The Wall Street Journal reported.
People made fun of Murati when she froze after being asked what Sora was trained on. But behavior like that indicates understanding that you could get the company sued if you said something incriminating. Altman just tweets through it.
[1] https://arstechnica.com/tech-policy/2024/05/openai-pauses-ch...
https://podcasts.apple.com/us/podcast/an-ex-cia-officer-expl...
https://nwn.blogs.com/nwn/2024/04/adam-schiff-ai-video-games...
This voice: https://x.com/OpenAI/status/1790072174117613963
As for why it's bad, it's because they set down the precedent of wanting specifically Scarlett Johansen's voice, got declined, doubled down, got declined again, and then went ahead and did it anyway. They can say in their own defense that it's some other voice actress that sounds similar, ok, so produce that name, tell us who she is.
Absent that, it's Johansen's voice, clipped from movies and interviews and shows and whatever.
An issue with voice actors having their voice stolen by AI models/voice cloning tech is that they have no legal standing because their performance is owned by their client, and therefore no ownership. ScarJo may not have standing, depending on the contract (I suspect hers is much different than typical VA). It might have to be Annapurna Pictures that sues OpenAI instead.
Forbes had a good story about performer rights of voices: https://www.forbes.com/sites/rashishrivastava/2023/10/09/kee...
IANAL of course.
Bette Midler was able to sue Ford Motor Co. for damages after they hired a sound-alike voice: https://en.wikipedia.org/wiki/Midler_v._Ford_Motor_Co. Ford had acquired the rights to the song (which Midler didn't write).
ChatGPT using Sky voice (not 4o - original release): https://youtu.be/JmxjluHaePw?t=129
Samantha from "Her" (voiced by ScarJo): https://youtu.be/GV01B5kVsC0?t=134
Rashida Jones Talking about herself https://youtu.be/iP-sK9uAKkM
I challenge anyone to leave prejudice at the door by describing each voice in totality first and seeing if your descriptions overlap entirely with others. They each have an obvious unique whispiness and huskiness to them.
When the offer was declined by scarjo, they could still train on her works of art and just hire a soundalike to make recordings regardless of whether they used it during training.
Then, at release time - either they get the buzz of artist-licensed "Her" or they get the buzz /outrage/Streisand of unlicensed "Her". Even if they take it down, OpenAI benefits.
I feel like the folks who fear the tech are wrong. But when the supposed stewards do such a moustache-twirling announcement, it seems like maybe we do need some restraint.
If a trade group can't put some kind of goodwill measures in place, we will inevitably end up with ham fisted legislation.
For me to believe this was genius I'd have to see some actual response from Sam. From the outside-looking-in, it appears that he was caught with his pants down when Jonhansson said no and went ahead even though he was rejected a second time and obviously knew it was the wrong choice. There's no Streisand effect at play here, OpenAI already owned the news cycle with their 4o announcement and could have kept it quiet. But Sam just had to have his One More Thing, and now he's getting his just deserts.
People cheering for this sort of copyright are completely lost imo. That's not a world anyone but the select few wants to live in.
Nevertheless that's not what happened here.
I mean I know he has hundreds of blind followers but good Lord, you would think that the man, with all his years of experience had some sense to introspect about what he is trying to achieve vs how he is going about it.
Money really does blind all our senses, doesn't it?
Sam doesn’t care.
After the board threw him out of his own company, why would he allow that to happen again? With that, he now trusts far much less people.
> Money really does blind all our senses, doesn't it?
That is why the cultishness was full on display last year when he was fired by the board.
mainstream adoption hasn't been that great - now there's drama
That said, the timeline she lays out is damning indeed.
All the while many people believe them at every step.
The top comment in this thread is crazy too, they probably contacted her two days prior to launch on the off chance that they could use her as a marketing puppet.
Lost for words on this one.
Hurray, OpenAI has found a new lucrative market. Horny incels.
[1] https://forums.theregister.com/forum/all/2024/05/21/scarlett...
They have similar voices, but SJ has more bass and rasp.
And if it's true that OpenAI hired a different actor, then this should basically be case closed.
The voice of Sky (assuming that's the same as the demo video), sounds like a run of the mill voice actor tbh. Great, but not that interesting or unique.
The real problem, now, is that they don't have a nice working voice anymore.
it doesn't seem like principles should matter. but then the bill of rights doesn't seem like it should matter either if you were to cold read the constitution (you might be like - hmm, kinda seems important maybe...).
it compounds culturally over time though. principles ^ time = culture.
"Audacious, Thoughtful, Unpretentious, Impact-driven, Collaborative, and Growth-oriented."
https://archive.is/wLOfC#selection-1095.112-1095.200
maybe "thoughtful" was the closest (and sam is apologetic and regretful and transparent - kudos to him for that). but it's not that clear without a core principle around responsibility. you need that imho to avoid losing trust.
OpenAI pulls Johansson soundalike Sky’s voice from ChatGPT - https://news.ycombinator.com/item?id=40414249 - May 2024 (96 comments)
And really, how much worse would the demo have been if they hadn't cloned Johansson's voice, and instead used another unknown voice? If it was similarly flirty, we'd have fallen for it anyways.
When is it infringing to make something that looks or sounds like somebody famous? I mean, there's only so many ways a human voice voice can sound or face can look. At what point are entire concepts locked down just because somebody famous exists or existed that pattern matches.
Doesn't matter around similarity. There was nothing fair-use around this voice and it is exactly why OpenAI yanked the voice and indirectly admitted to cloning her voice.
Their efforts to copy the Her character mannerisms is still annoying. They were aiming for the Her-like personality. But it's a stretch to say it's a copy of Johansson's voice. The haze in her real voice isn't there in GPT. Maybe they decided to pull the voice because it's not good publicity to have Scarlett Johansson pissed off with you.
Nevertheless these tech companies should hire professional film writers and artists to help with difficult concepts such as original ideas and not copying other's work.
I don't know guys, the super hyped up company with next-gen technology might just be using crime, underhanded tactics, and overstating their capabilities to pull in the thing we all love... and it's not each other or your friend's mother!
It's money!
Assuming Sam Altman is not stupid, this could be part of some elaborate plan and a calculated strategy. The end goals could range from immediate practical outcomes like increased publicity (see ChatGPT's mobile app revenue doubled overnight: https://finance.yahoo.com/news/chatgpts-mobile-app-revenue-s...) and market impact, to more complex objectives like influencing future legal frameworks and societal norms around AI.
It's an excellent book, and so so many of the issues raised in it are playing out blow-by-blow.
To me, that reads like the same kind of snake oil he sold Elon when he proposed the joint founding of OpenAI.
I can just about imagine the books in his private library. The Prince. 48 Laws of Power. Win Friends and Inference People.
Very interesting to see this there. Does anyone know how could that be legislated?
Given that connection, I think it's plausible. They have a similar voice and given Bari's experience in podcasting could be a sensible choice if OpenAI wanted Scarlett but couldn't make it happen.
>"I don't know about the voice. I actually had to go and listen to Scarlett Johansson's voice," Murati said.
Seems like a big part of Mira’s job is not knowing things. How is no one questioning how she landed a VP job at OpenAI 2 years after being an L5 PM?
I assume that they didn't get Scarlett's voice directly but they hired someone who sounds similar to use it for the system.
Is it illegal to hire similar sounding voice actor?
If I, for example, sound very similar to Stephen Fry, would I not be allowed to record audio books because he _owns_ his voice and any similar voice?
Moreover, what if some actor has a similar voice to her and records and sells an audiobook? What if this actor signs a contract with OpenAI?
Problem solved.
For example, I would love to see all of the Bourne books adapted into live-action films, but I know that will be impossible. In the future, I believe it would be great to see some AI actors who are not related to any famous actors/actresses perform the same screenplay: of course, if the book is licensed to that AI movie.
I'll ask the devil's advocate / contrarian question: How big a slice of the human voice space does Scarlett lay a claim to?
The evidence would be in her favor in a civil court case. OTOH, a less famous woman's claim that any given synthesized voice sounds like hers would probably fail.
Contrast this with copyrighted fiction. That space is dimensionally much bigger. If you're not deliberately trying to copy some work, it's very unlikely that you'll get in trouble accidentally.
The closest comparison is the Marvin Gaye estate's case. Arguably, the estate laid claim to a large fraction of what is otherwise a dimensionally large space. https://en.wikipedia.org/wiki/Pharrell_Williams_v._Bridgepor...
Celebrities need to get used to the fact that they will soon be no more important to a corporation than any other rank and file employee. AI is already able to conjure up whatever voices and personas we need at the ready and make the concept of actors all but a thing of the past.
It’s interesting to see how it unfolding.
Altman and Murati are world-class grifters but until now they were stealing from print media and digital artists. Now they’re clashing with some of the most litigious industries with the deepest pockets. They’re not going to win this one.
Who would have thought we would be discussing voice theft someday.
move fast break people
Imitating a movie AI was a cool idea and imitation was the only legal way to do it.
Do you pull your hair when companies advertise with Elvis impersonators?
Nobody was significantly harmed by this, I can guarantee the rich people that use hacker news consume things from much less savory standards than imitating a celebrity.
Nestlé is strong but you pull the plug at THIS?
Pg has done worse and he owns this forum.
Have some perspective.
The worst of it is not that this one person is being ripped off (that's bad enough and I hope she gets some kind of resolution). The worst of it is that it shows the company and the people behind it who are making the big decisions are dishonest and unethical.
All the alleged "safety" experts in corporations and in government policy and regulators? All bullshit. The right way to read any of these "safety" laws and policies and regulators is that they are about ensuring the safety of the ruling class.
Maybe Altman lands in jail or files for bankruptcy after all the dust settles.
If that box wasn’t in your bingo card I’m sorry, it’s basically the center/free box at this point.
In the not so distant future, when the world's top AI models can generate endless accents and voices at will, the probability of one of those sounding just like you (and thousands of other people) will be high. It will be VERY high.
All this dealing with Hollywood and music industry and all the crap i've been reading about OpenAI trying to wiggle their way into those industries is absolute damn nonsense. What is Sama thinking?! GO BACK TO BEING NERDS AND STFU.
If you really believe you are going to create a real AGI, none of this is relevant. No one is going to thank you for creating something that can replicate what they value in seconds. Do it anyway.
And remember, STFU.