OpenAI announced on December 9 that its AI video-generation tool, Sora, would launch later in the day. The tool can generate video clips inspired by still images, extend existing videos, or fill in missing frames.
In a YouTube livestream, the company revealed that Sora will debut to U.S. users and most countries internationally. However, the firm admitted it has no timeline for launching the tool in Europe, the UK, and some other countries.
OpenAI seeks to balance Sora with creative expression
The company revealed that Sora will work similarly to OpenAI’s image-generation AI tool, DALL-E. Users will be required to type out a desired scene, and Sora will return a high-definition video clip.
OpenAI introduced Sora in February after the company burst into the mainstream last year, thanks to ChatGPT’s viral popularity. The firm revealed that its users will not need to pay extra for the tool because it will be included in existing ChatGPT accounts such as Plus and Pro.
The company’s CEO, Sam Altman, revealed features like Blend during the livestream. It emphasized that the feature can join two scenes together at the user’s direction. Sam Altman also disclosed that Sora will have an option to make an AI-generated video loop endlessly.
Users questioned OpenAI executives during a Reddit AMA in October about a potential delay in Sora’s release date “due to the amount of compute/time required for interference or due to safety.”
The company’s product chief, Kevin Weil, said it needed to perfect the model and get safety, impersonation, and other things right. Weil also added that the company needed to scale its computing power.
“We obviously have a big target on our backs as OpenAI, so we want to prevent legal activity of Sora, but we also want to balance that with creative expression.”
-Rohan Sahai, OpenAI’s Sora product lead.
The firm also highlighted that Sora has mainly been available to a small group of safety testers who test the model for vulnerability in areas such as misinformation and bias.
Sora wants to pioneer the generative AI market
The Microsoft-backed artificial intelligence startup closed its latest funding round in October at a valuation of $157 billion. The firm hopes to get a slice of the generative AI market as it battles with companies like Amazon-backed Create with Alexa. The generative AI space has been predicted to top $1 trillion in revenue within a decade.
The company hired its first chief marketing officer earlier this month, revealing its plans to spend more on marketing to grow its user base. It also launched a search feature within ChatGPT that could potentially put it in a position to compete with search engines like Google’s Veo.
Sora’s launch comes amid a recent protestors’ leak that appeared to be a copy of Sora over concerns about the ChatGPT maker’s treatment of artists. Some members of OpenAI’s early access program for Sora published an open letter in late November critiquing the firm for not being sufficiently open to supporting the arts beyond marketing.
However, the company’s spokesperson replied in a statement to CNBC that participation was voluntary, and there was no obligation to provide feedback or use the tool. The spokesperson added that the company was excited to offer artists free access to Sora and would continue supporting them through grants, events, and other programs.
From Zero to Web3 Pro: Your 90-Day Career Launch Plan