Runway plans to make Gen-3 Alpha available to customers soon, using it to power three cloud services for video and image generation from text and user prompts. The company also plans to offer customized versions of the model to enterprises, allowing them to align the style of AI-generated videos with their project requirements. The company's long-term AI development roadmap focuses on general world models, simulating objects based on complex data and "realistic models of human behavior".
Key takeaways:
- Startup Runway AI Inc. has launched a new artificial intelligence model, Gen-3 Alpha, that can generate ten-second videos based on text prompts.
- Gen-3 Alpha is capable of generating a 10-second clip in 90 seconds and is designed to produce higher-fidelity videos with better depiction of motion and frame consistency.
- Runway is developing safety features for Gen-3 Alpha to prevent its misuse, including a provenance system based on C2PA standard that will indicate if a video was AI-generated.
- The company plans to make Gen-3 Alpha available for customers soon, and will use it to power three cloud services for video and image generation from text and user prompts.