Nvidia-backed $4 billion AI startup announces major London expansion

Luma AI, a video generation startup backed by Nvidia, is set to make significant strides in the U.K. tech scene with its recent announcement of a major expansion in London. The Palo Alto-based company plans to recruit approximately 200 new employees, which will represent nearly 40% of its total workforce, by early 2027. This new London hub will focus on various key areas, including research, engineering, partnerships, and strategic development. This expansion follows Luma's impressive funding round earlier this month, where it secured $900 million led by Humain, an AI firm backed by the Saudi Public Investment Fund. This funding has propelled Luma's valuation to over $4 billion. The startup is renowned for developing "world models," advanced AI systems that can learn from video, audio, images, and text, similar to the large language models (LLMs) that power platforms like OpenAI's ChatGPT and Google's Gemini. Targeting sectors such as marketing, advertising, media, and entertainment, Luma's video models are accessible through an API and are integrated into a comprehensive content creation suite. Amit Jain, CEO and co-founder of Luma AI, expressed excitement about the expansion, stating, "With this Series C raise and the upcoming build-out of global compute infrastructure, we have the capital and capacity to bring world-scale AI to creatives everywhere." He views the move into Europe and the Middle East as a natural progression to empower storytellers and brands worldwide. Jain highlighted the U.K. as a strategic starting point for this expansion, attributing its appeal to the abundance of talent available, particularly in London, where prestigious universities and institutions like DeepMind thrive. He mentioned, "London has some of the best people when it comes to research." The city is also considered a gateway to the European market. Luma AI joins a growing list of North American AI companies expanding into the U.K. and Europe to leverage local talent and business opportunities. Notable firms like Anthropic, Cohere, and OpenAI have also announced plans to establish offices across Europe in recent months. While world models are still evolving, some researchers believe they are critical in the quest for artificial general intelligence (AGI). Jain noted that these visual models are about a year to a year-and-a-half behind language models, but he anticipates they will soon become the primary interface for AI in everyday applications, given the increasing time people spend consuming video content. Major tech players, including Google, Meta, and Nvidia, are also exploring world models for diverse applications. Luma AI recently launched its latest model, Ray3, which Jain claims outperforms OpenAI's Sora and performs comparably to Google's Veo 3.