1. OpenAI’s generative video tool Sora is set to officially launch later this year, with the goal of making it more efficient and cheaper before release.
2. The AI video generator is being tested to ensure it does not produce unsavory or misleading content, with plans to implement safety measures to prevent misinformation.
3. Sora will have limitations on release, similar to DALL-E’s guardrails, to prevent illegal, unrealistic, or potentially libelous content from being generated.
OpenAI’s generative video tool, Sora, is set to be officially launched later this year, with the goal of making it more efficient and affordable before its release. The tool will be capable of producing high-quality videos from text, potentially being integrated into popular video editing tools like Adobe Premiere Pro or Apple Final Cut Pro.
A recent survey found that most adults were unable to distinguish between videos created by Sora and real footage, highlighting the need for improved identification measures to prevent misinformation. OpenAI is addressing this issue by conducting red teaming exercises to ensure the tool does not produce harmful or misleading content.
While Sora is still being refined to improve speed and functionality, OpenAI is also considering the ethical implications of its release, particularly in relation to global events like elections. The company is working with artists and policymakers to determine boundaries regarding sensitive content like nudity and violence, with the aim of making Sora a tool for enhancing creativity rather than causing harm.
Mira Murati, OpenAI CTO, emphasized the importance of ensuring the responsible deployment of AI tools like Sora, acknowledging the challenges that come with integrating such advanced technology into everyday use. Despite these challenges, she believes the benefits of extending creativity through AI tools are worth pursuing.