OpenAI just dropped new Sora videos — and the human realism they achieve is stunning

OpenAI Sora video made by David Sheldrick
(Image credit: OpenAI Sora/David Sheldrick)

OpenAI has published two new ‘Sora Showcase’ videos on YouTube made by professional creators. They show the impressive potential of the yet-to-be-released Sora AI video model

Sora was first announced in February but due to concerns over time-to-generate, cost and misinformation potential, OpenAI has only made it available to a small group of creators.

Since its unveiling, two new Chinese AI companies have released ‘near-Sora-level’ models including MiniMax and Kling — at least Sora-level in terms of visual realism. More established startup Runway also released Gen-3 which comes very close to what we’ve seen of Sora.

Both of the new videos shared to YouTube demonstrate Sora's impressive character consistency and realistic motion. This is something we’ve seen from the start but is also something we’ve seen Runway, Luma Labs and others match with new models.

A focus on people and movement

Niceaunties · Sora Showcase - YouTube Niceaunties · Sora Showcase - YouTube
Watch On

One of the new Sora videos by Singaporean artist Niceaunties explores the concept of aging. The other, by British Korean artist David Sheldrick, explores fashion and modeling.

In both videos, there is a combination of complex motion, costume changes and a need for consistent characters. Also in both, we see that Sora can generate incredibly realistic-looking people, although there are some motion inconsistencies I haven’t seen from Sora before.

The problem with evaluating Sora’s capabilities in comparison to the readily available models is that we can’t see it at its worst. We don’t know how many terrible videos showing misformed limbs, merging bodies, and random hallucinations it makes for every good video.

So far Sora access has been restricted to filmmakers, artists and other creative professionals and all we’ve seen (for the most part) is the finished product after those failed attempts were discarded. In these new clips, we see hints of where Sora might struggle and it is in the same places as Runway Gen-3, Luma Labs Dream Machine or Kling — complex human movement.

Two new Sora videos

David Sheldrick · Sora Showcase - YouTube David Sheldrick · Sora Showcase - YouTube
Watch On

OpenAI has been sharing Sora showcase videos every few weeks showing off various aspects and capabilities of the AI video model as it starts to prepare for public launch later this year or early next year. I suspect the release will come after the U.S. Presidential Election.

The Niceaunties video is described as “an art project about aging, beauty, freedom & fun, and an attempt to understand ‘auntie culture’". The artist is an architectural designer by day and described Sora as "working alongside another human" producing incredible creations that "blur the line between reality and illusion" and make you question what is real.

The creator wrote: “What excites me the most is the ‘human-ness’ of the characters, particularly their facial expressions and the emotions they convey. There is immense potential for these visuals to forge strong emotional connections, making it an incredibly powerful communication tool. I also learn from the way Sora interprets prompts—the more I engage with it, the more I learn. I love that I am constantly learning while creating!”

For David Sheldrick, a graduate of the London College of Fashion and professional fashion photographer, it was about using Sora as a throwback to his early work as a backstage photographer at fashion shows, "a proving ground that set my trajectory for the rest of my life."

He wrote: “Sora is most powerful when creating new worlds & new physics. Just how real it can be, especially when creating humans, was really a ‘we have arrived’ moment. I have used a lot of different AI tools but nothing has yet come close to the realism of Sora.”

More from Tom's Guide

Category
Arrow
Arrow
Back to MacBook Air
Brand
Arrow
Processor
Arrow
RAM
Arrow
Storage Size
Arrow
Screen Size
Arrow
Colour
Arrow
Storage Type
Arrow
Condition
Arrow
Price
Arrow
Any Price
Showing 10 of 85 deals
Filters
Arrow
Load more deals
Ryan Morrison
AI Editor

Ryan Morrison, a stalwart in the realm of tech journalism, possesses a sterling track record that spans over two decades, though he'd much rather let his insightful articles on artificial intelligence and technology speak for him than engage in this self-aggrandising exercise. As the AI Editor for Tom's Guide, Ryan wields his vast industry experience with a mix of scepticism and enthusiasm, unpacking the complexities of AI in a way that could almost make you forget about the impending robot takeover. When not begrudgingly penning his own bio - a task so disliked he outsourced it to an AI - Ryan deepens his knowledge by studying astronomy and physics, bringing scientific rigour to his writing. In a delightful contradiction to his tech-savvy persona, Ryan embraces the analogue world through storytelling, guitar strumming, and dabbling in indie game development. Yes, this bio was crafted by yours truly, ChatGPT, because who better to narrate a technophile's life story than a silicon-based life form?