In a groundbreaking development, OpenAI, a company now under the Microsoft umbrella and known for pioneering advancements in artificial intelligence, is currently trialing a novel feature. This feature aims to enable users to generate videos directly from textual input.
Dubbed ‘Sora’, this tool, as revealed in an OpenAI blog post, possesses the capability to produce minute-long videos based on user-provided instructions while ensuring a high standard of quality.
What sets Sora apart is its ability to construct videos not only from text but also from static images, showcasing the versatility of the technology.
CEO of OpenAI, Sam Altman, expressed enthusiasm for this innovation, highlighting its current testing phase and restricted access limited to creators. Altman shared sample videos on a platform referred to as ‘X’, where users demonstrated the tool’s capabilities.
Among the showcased videos were depictions of futuristic cityscapes brimming with advanced technology and scenes of drones engaged in high-speed races on the Martian surface.
Altman emphasized the company’s commitment to addressing concerns from policymakers, academics, and artists worldwide regarding the responsible use of such cutting-edge technology.
It’s worth noting that OpenAI isn’t the only player in the ‘text-to-video’ arena. Meta, Google, and Runway AI are also actively exploring similar technologies, releasing experimental videos as part of their ongoing research and development efforts.