Sora - OpenAI's text to video generator

[Paul_Heery]

OpenAI has introduced Sora, their text-to-video model.

Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.

https://openai.com/sora

There are dozens of examples on the page that were all generated with just text prompts.

Sora is out and available now, although it is currently only usable by red teamers to assess critical areas for harm or risks. OpenAI is also granting access to a select group of visual artists, designers, and filmmakers to give feedback on advancing the model even further.

https://openai.com/sora?video=cat-on-bed

[Arizona_928]

Glad to see my 20 bucks a month go to use

[GH85Carrera]

I suspect than in the very near future, there will be a move at the theaters that is 100% AI, and no human actors on screen. They may need them for voice for a while but it will be a new category for the Academy awards. No one but the programmers to accept the awards.

So many movies already are mostly computer generated effects, just an actors face and voice. Not yet, but before the end of the decade. No more expensive Hollywood union paid crew, or actors. And no more actor scandals, or crazy demands. Just all digital.

[jyl]

Lots of implications.

No more need for actors, directors, crews, props, costumes, locations - basically all of the global film and TV industry goes away, except for the writers and marketers, but they go away too.

No more need for photograpers, videographers, stock footage, video or sound editors -basically all of that industry goes away.

No more need for game coders, and videogames is a bigger industry than the whole film and TV industry.

Okay, where labor is strong, they might be able to slow this - stuff that isn’t needed might still be required, by union contact or laws . . . but that won’t last, nothing can long stand in the way of money.

No more need for artists either, by the way. Why make any image with hands on brush, charcoal, or stylus, when it can simply be generated? Sure, a small number of conoisseurs won’t accept AI generated art, then they’ll age out and take their objections with them.

That’s just some of the commercial implications. The social implications are worse.

There will be no way to know if a photo or video are factual or invented. Period. Did you know that the NYPD murdered an entire family, raped the women and eviscerated the men, shot the children and burned down their house, last week? Here’s the video and audio. Here’s video proof. Did you know that the Governor of Texas was caught screwing underage models in bondage? Here’s proof. Did you know that gunmen from Gaza took over a music festival in Israel and politely greeted the Jewish kids before leaving them unharmed? Here’s proof. Did you know that Auschwitz was a voluntary stay center where refugees from Allied bombing received shelter, medical care and food and children went to school? Proof.

Laughable? Not in a country where a fifth of people think Taylor Swift and Travis Kelce are part of a Pentagon plot to steal the election. Or where more than a fifth of people think the Holocaust never happened, like the moon landings. And if you think people are stupid now, wait until a generation of kids grows to adulthood having never seen or heard anything but invented video and audio.

You’d think Congress would be all over this. Mandatory watermarking, disclosures, no AI images of actual people or hostorical events, etc - pretty obvious stuff that can be done. But money talks, and there is now more money behind AI than any industry in the history of the world.

[sc_rufctr]

[tdw28210]

AI generated movie trailers have been around for awhile now. At least a year in the wild. However, they've not been made via a strictly voice to text method. They've required multiple AI tools. Most of them look amazing.