OpenAI said on Monday it would release Sora, its popular artificial intelligence video generation tool, later in the day.
The AI video generation model works similarly to OpenAI’s image generation AI tool DALL-E: the user inputs the desired scene, and Sora will return high-definition video clips. Sora can also produce movie clips inspired by still images and extend existing footage or fill in missing frames. The Microsoft-backed artificial intelligence startup, which went mainstream last year with the viral popularity of ChatGPT, launched Sora in February.
According to OpenAI’s YouTube livestream, the tool will first launch to users in the United States and “most countries internationally” later today, and the company has not yet “determined a timetable for launching the tool in Europe, the United Kingdom, and some other countries.”
OpenAI says users will not have to pay extra for the tool, which will be included with existing ChatGPT accounts such as Plus and Pro. Employees on the livestream and OpenAI CEO Sam Altman demonstrated features such as “blending” (that is, joining two scenes together according to the user’s instructions), and the option to make the AI-generated video repeat indefinitely.
Until now, Sora has been primarily used by a small group of security testers, or “red teamers,” who test models for vulnerabilities in areas such as misinformation and bias.
Reddit users asked OpenAI executives in October about Sora’s release date, questioning whether its delay was “due to the amount of computation/time required for inference or security reasons.” In response, OpenAI’s product lead Kevin Weil wrote: “Models need to be refined, safety/simulation/other things need to be correct, and computation needs to be scaled!”
Rohan Sahai, OpenAI’s Sora product lead, said during the live broadcast: “As OpenAI, we obviously have a big goal, so we want to prevent Sora’s legal activities, but we also want to balance that with creative expression.”
OpenAI completed its latest funding round in October at a valuation of $157 billion, including $6.6 billion the company raised from numerous investment firms and large technology companies. It also secured a $4 billion revolving credit facility, bringing its total liquidity to more than $10 billion.
This is all part of OpenAI’s serious development plan, because Microsoft-Support the battle for artificial intelligence startups Amazon-Supports Anthropic, Elon Musk’s xAI, Google, YuanMicrosoft and Amazon occupy the largest share of the generative artificial intelligence market, Expected to exceed US$1 trillion Income over ten years.
Earlier this month, OpenAI hired its first chief marketing officer, signaling plans to spend more on marketing to grow its user base. In October, OpenAI launched a search feature in ChatGPT, allowing it to better compete with search engines such as Google, Microsoft’s Bing and Perplexity, and potentially attract more users who visit these sites for web searches.
ChatGPT maker hopes to use Sora to compete with video-generating artificial intelligence tools from the likes of Meta and Google, which announced the launch of Lumiere January. Other startups offer similar AI tools, such as Stability AI’s Stable Video Diffusion. Amazon also announced Create with Alexa, a model specifically designed to generate prompt-based short-form animated children’s content.
Now that chatbots and image generators have entered the consumer and business worlds, video may become the next frontier of generative artificial intelligence. While the creative opportunities will excite some AI enthusiasts, the new technology raises serious misinformation concerns as major political elections take place around the world. Data from machine learning company Clarity shows that the number of deep fakes generated by artificial intelligence has increased by 900% year-on-year.
OpenAI has made multimodality (a combination of text, image and video generation) a prominent goal in its efforts to provide a broader suite of artificial intelligence models.
News of Sora’s release comes after protesters decided to leak what appeared to be a copy of Sora due to concerns about the way artists were treated by the makers of ChatGPT.
Some members of OpenAI’s Sora early access program, which reportedly includes about 300 artists, released a open letter In late November, OpenAI was criticized for not being open enough or supporting art other than marketing.
The protesters’ open letter reads: “Dear Corporate AI Overlords, We licensed Sora and promised to be early testers, red teamers, and creative partners. However, we believe we are being lured into ‘art Cleanse’ comes to tell us that the world knows Sora as a useful tool for artists.
The letter adds that hundreds of artists have provided pro bono labor to OpenAI through bug testing and feedback on Sora, and that “while hundreds of artists contribute for free, a select few will be selected through a competition to screen their Sora-created films – Offering minimal compensation pales in comparison to the tremendous PR and marketing value OpenAI receives.
“We are not opposed to the use of artificial intelligence technology as an artistic tool (if we were, we probably would not have been invited to participate in this project),” the open letter said. “What we disagree with is how this artist project was launched and how it was launched. How the tool was formed ahead of a possible public release. We share this with the world in the hope that OpenAI becomes more open, artist-friendly, and supports art beyond PR stunts.
In late November, an OpenAI spokesperson responded to the protesters’ actions in a statement to CNBC.
“The hundreds of artists in our alpha shape the development of Sora, helping prioritize new features and safeguards,” an OpenAI spokesperson said at the time. “Participation is voluntary and there is no obligation to provide feedback or use the tool. We are We are pleased to provide free access to these artists and will continue to support them through grants, events and other programs.”
About The Author
