Runway AI, the creator of Gen-1 and the co-creator of Stable Diffusion, has announced the launch of another model in its generative AI hat called Gen-2. The new AI model improves upon its predecessor Gen-1, a text-to-video AI model that relied on users to improve existing videos by going a step further and creating videos from plain text.
With their Gen-1 AI model launched at the beginning of February, Runway continues its research of diffusion models into an image and video synthesis. Gen-1 users take existing videos and images to overlay an AI-generated layer. The new videos can be used to create animated films and various projects. Runway AI also helped initially to create Stable Diffusion, a text-to-image AI model.
Gen-2 brings forth a whole new frontier, any text that comes to your mind can be converted into a video with their new AI model. Not only will Gen-2 create full-feature videos from your text prompts, but it will also use images and video clips like Gen-1 to create fresh films.
Although not yet launched, Gen-2 has already started a waitlist on Runway’s Discord community to sign up for Gen-2.
Inducing creativity in society, Gen-2 generative capabilities will provide users around the world with quick videos without investing time and money into learning the tedious video-editing process. Not to overshadow anyone who enjoys this process, Gen-2 will be a worthy partner for any filmmaker to edit and piece AI-generated videos together from just text prompts.
With only one video in their arsenal and that too with clips from Gen-1, Gen-2 still has a long way to go before it becomes precise and realistic. In time, Gen-2 will join generative AI tools to eventually shape itself to the demands of the prompts put forth by users.
Is Gen-2 going to blur the lines between real in AI-generated?
While it can be looked upon as a tool to help humankind explore and indulge in creativity, it might become harder to differentiate between AI-generated and human-produced artworks, be they films, videos, photos, or paintings, among others.
Although many images or clips might be blurry, Runway is continually competing with Midjourney and Dall-E to provide the best text-to-image generative AI tool.
With an AI generating a video or an image from a few prompts, artists globally are bound to feel insecure if their clients choose AI to synthesize a video or an image according to their own ideas rather than approaching an artist to visualise the same. However, it can also be seen as an opportunity for artists to upskill their work and incorporate AI into their portfolio to attract clients.
In the News: Punjab internet shutdown extended; Amritpal Singh on-the-run