Rhoda hits unicorn status with $450M round
Video-first AI startup Rhoda has raised $450 million in fresh funding at a $1.7 billion valuation, underscoring investor conviction that the next wave of artificial intelligence will be trained not just on text, but on video. The latest round is backed by prominent venture capitalist Vinod Khosla and his firm Khosla Ventures, positioning Rhoda among the fastest-growing players in the emerging video-native AI segment.
A new generation of video-trained AI models
Rhoda is developing large-scale AI models that learn directly from video, rather than relying primarily on text or still images. By ingesting vast libraries of moving images, the company aims to build systems capable of understanding motion, context, cause-and-effect and real-world dynamics in ways that conventional text-based large language models cannot fully capture.
The startup’s technology is designed to power applications ranging from autonomous robotics and smart surveillance to media automation and digital content creation. Investors see this video-first approach as a critical next step in AI, enabling machines to interpret the physical world with far greater nuance.
Strategic backing from Khosla Ventures
Khosla Ventures, known for early bets on transformative technologies, has framed video-trained AI as a foundational shift similar to the rise of deep learning a decade ago. The firm’s support is expected to give Rhoda both capital and strategic guidance as it scales its infrastructure, builds out its research team and competes with larger AI labs.
Scaling infrastructure and commercial use cases
The new capital will be used to expand GPU-intensive compute infrastructure, grow engineering and research headcount, and accelerate the rollout of commercial products. Rhoda is targeting enterprise customers that need advanced computer vision and multimodal AI capabilities, including sectors such as security, industrial automation, entertainment and advertising.
As regulators and customers scrutinise how training data is sourced and used, Rhoda is also expected to invest in data governance, privacy safeguards and responsible AI frameworks to ensure compliant and ethical deployment of its video-trained models.