OpenAICTO Meera Murthy reveals AI text-to-video generator Sora will be available later this year: Report

Open AI It plans to introduce the latest generation. Artificial intelligence (AI) model Sora later this year, according to a report. Sora, introduced Last month, there was an AI-powered text-to-video generator that the company claims can create videos up to 60 seconds long, far longer than any of its industry rivals. In a recent interview, the AI ​​firm’s chief technical officer also revealed that Sora can instantly create videos in minutes depending on the complexity. The interview also covered topics such as model error, the data used to train it, and the potential value of Sora.

In one The interview Asked when the AI ​​video generator would be available, OpenAI CTO Meera Murthy said to the Wall Street Journal, “I’m definitely hoping this year, but it could be a few months.” Asked if the US presidential election, which is scheduled to take place in November 2024, will factor into the company’s thoughts on when to release an AI model, Marathi said tackling misinformation and harmful bias is important for the company. And added, “We will not release anything that we do not believe will affect the global elections.”

Murthy was also asked about the data source on which Sora was trained. The question is also his. interest Italy’s data protection watchdog is investigating potential data breaches by OpenAI. Regulators gave the AI ​​firm 20 days to provide information about the AI ​​video generator.

In the interview, Marathi also did not reveal any details and only said that the AI ​​model was trained on data available in the public domain and licensed to the company. She also said she wasn’t sure whether YouTube or Facebook data was used, but confirmed to the publication that Sora used content from Shutterstock, with which they partner. Is.

Other OpenAI tools such as Sora are also a very expensive model to run. Chat GPT And DALL-E, Murthy explained. “ChatGPT and DALL-E are optimized for public use while Sora is really a research output. It’s much more expensive,” the CTO told the WSJ. While the company hasn’t set a price point for the AI ​​Video model, Marathi said the company is looking to make the model available, once it’s released, at the same price as the DALL-E. Currently, the base price of the DALL-E 3 model is $0.040 (roughly Rs. 3) per image at 1,024 x 1,024 pixels resolution.

Further, the OpenAI CTO also highlighted that the company is working on further improvements to bring audio capabilities to Sora, improve generation time and its cost-effectiveness, and reduce errors.

---