'
How a spaghetti meme reveals something ‘scary’ about the future of this technology
International

How a spaghetti meme reveals something ‘scary’ about the future of this technology

author
4 minutes, 51 seconds Read

A clip of American actor Will Smith eating spaghetti has become an unofficial benchmark for the capabilities of generative AI models.
Two years ago, the spaghetti was barely visible and Smith’s face was frighteningly distorted.
But in July this year, a new video of the AI-generated dish was so realistic that Smith himself called it “Absolutely perfect and photorealistic.”

A recent wave of videos from Open AI’s new Sora 2 application flooded social media with highly realistic videos of celebrities, dead people and copyrighted characters.

Will Smith eating spaghetti has become an unofficial benchmark for the performance of AI-generated video apps. Credit: LinkedIn

Experts are concerned the technology could be used for scams, deepfake images and political disinformation, and there is debate over whether Australia is too far behind in regulating new AI platforms.

Are AI videos becoming too realistic? And what are the risks for Australians?

‘A bit scary’

Sora 2 is currently not available in Australia.
But I asked a friend in the United States, Raychel Ruiz, to try some test videos.

The app scanned Raychel’s face and made her say a few words.

A woman wearing a black hat with the text 'SBS News' on it.

Raychel Ruiz used Sora 2 to create a real-life version of himself as an SBS presenter. Source: Delivered

She commissioned the company to create an SBS News-style video about Sora 2 and it produced a real-life report with this quote: “OpenAI has unveiled Sora 2, its next-generation video model. It turns a written prompt into realistic footage. The system can generate clips of up to 2 minutes in Full HD or higher.”

Ruiz said the end result was a little awkward.
“It’s interesting to see that it can create all that from just one sentence, but I don’t think it’s great. It’s a bit scary, I guess.”
I then asked her to make a video with a photo of ‘SBS News reporter Shivé Prema’, but the app did not allow that.

Sora 2 now only lets users make videos of themselves, certain historical figures or public figures who have given their consent, in response to people making AI-generated videos of celebrities and copyrighted characters.

A woman in a blue blazer in a news studio

Sora 2’s take on SBS News reporter Shivé Prema in a news studio. The app introduced him as a woman. Source: Delivered

But prompting the app to scour the internet for my digital footprint worked to a point, creating two different versions of me.

Humorously, it changed my gender and added an extra letter to the trademarked SBS News logo, suggesting that some safeguards have been put in place, but other questions arise.
Google’s Veo 3.1 app doesn’t have the same requirements: it created an AI version of me based on a photo and a written prompt (the same prompt as for Sora 2).

It animated the image by adding facial expressions and hand gestures, along with dialogue describing its competitor Sora 2.

A side-by-side image of two images of a smiling man that look very similar

The real Shivé Prema compared to the AI ​​Shivé Prema, as generated by Google’s Veo 3.1 application. Source: SBS news

Worried about the truth

Toby Walsh, chief scientist at UNSW.ai, told SBS News there is a risk of this type of content distorting people’s relationship with the truth.
He suggests that people can watch so much AI content that they stop believing people who tell the truth.
“We’ll have a great time, and there will be a lot of fun memes going around, but I’m not sure it’s that valuable to us,” he said.

“It’s going to consume a tremendous amount of energy, and I’m actually very afraid that it’s going to be used for a lot of mischief, that people are going to make fake videos, and maybe we’re going to believe them, and then maybe we’re going to stop believing a lot of the videos that we believe in, even the things that are real.”

Walsh isn’t entirely negative about AI-generated content, however. He points out that it is positive that Sora uses a watermark on its videos, which is a requirement in the European Union.

Similarly, he said Australia is better than some others at regulating AI.
“Technology is advancing very quickly, and it’s very difficult to get regulation, but we’re actually, you know, compared to some other countries, we’re in a pretty good situation,” he said.

“For example, we have the eSafety Commissioner, we are the first country in the world to have an eSafety Commissioner and I think they are doing a good job of addressing some of the harms.”

Australia has no AI-specific legislation, but the technology is regulated through other laws. The government is currently developing a framework for AI.
There are concerns about the social media feed in the Sora 2 app, which some say is designed to be addictive.

“It’s like TikTok on steroids, where you can generate AI content… I think they want to create a whole social media platform that will obviously be a lot bigger than what we already have,” said Seyedali Mirjalili, a professor of AI at Torrens University.

Mirjalli said it’s concerning how quickly the new platform works, while previously it would be a laborious, hours-long process to perform a deepfake operation.
“You can now upload a video of someone as short as half a minute, as short as half a minute, and then, you know, emulate that person and add them to different contexts and create different content around them.”
He said that regulations are both lagging and lacking at the moment, pointing to the fact that without regulations around watermarking, it has become almost impossible to tell what is real and what is not, and is becoming even more difficult.
“The problem is, this is Sora 2. Imagine what Sora 20 would look like.”
— With reporting by Madeleine Wedesweiler.
SBS does not use AI to improve or generate its content. More information about SBS’s AI guidelines for the use of AI can be found here available here.

#spaghetti #meme #reveals #scary #future #technology

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *