AMI Labs, a European AI research institute, has something real to show for their under-the-radar work—it’s raised an incredible €890 million (around $1.03 billion). This funding will enable them to accelerate their vision to create world models that truly understand the nuances of reality. The organization is headquartered in Paris with major offices in New York, Montreal, and Singapore. This unique strategic positioning sees it leading the pack in AI innovation.
AMI Labs was founded by Yann LeCun, who serves as chairman and remains actively involved. The nonprofit is committed to continuing to do original, basic research rather than simply pursuing commercial applications. The organization has an amazing team to support their efforts. These members include Laurent Solly, Chief Operating Officer, Saining Xie, Chief Science Officer, Pascale Fung Chief Research & Innovation Officer and Michael Rabbat, Vice President of World Models.
That last funding round was highly competitive, with many investors scrambling for a piece of the company. This excitement is a testament to the ambition and potential of AMI Labs. Alexandre LeBrun, co-lead and chief architect of the project, about the immense scope and vision of the project.
“AMI Labs is a very ambitious project, because it starts with fundamental research. It’s not your typical applied AI startup that can release a product in three months, have revenue in six months and make $10 million in [annual recurring revenue] in 12 months.” – Alexandre LeBrun
AMI Labs has raised a total of $12 million. At the same time, which might be taking the world by storm raising the most talked about $1 billion of capital. This influx of capital reflects a growing interest in AI research and development, particularly in understanding and modeling the world around us.
Another new player in the AI fund space, SpAItial, just closed a $13 million seed round. That’s reflective of a larger movement, where record investments are flooding into cutting-edge AI startups. There’s no doubt that these companies are committed to testing the limits of technology.
AMI Labs deeply believes in the importance of transparency and collaboration within the AI community. Open-source initiatives, LeBrun explained, were foundational to their overall operational strategy.
“We will also make a lot of code open source,” he stated. “We think things move faster when they’re open, and it’s in our best interest to build a community and a research ecosystem around us.”
This principle aligns with AMI Labs’ mission to build enriched world models. To actually learn well, though, they need the honest real-world data and independent evaluations we provide. Research conducted solely in a lab does not allow for a full understanding of complicated systems and the team is committed to going beyond those walls.
“We are developing world models that seek to understand the world, and you can’t do that locked up in a lab. At some point, we need to put the model in a real-world situation with real data and real evaluations.” – Alexandre LeBrun
One of the most exciting new ideas in the field came from Yann LeCun’s lab, and it was the Joint Embedding Predictive Architecture (JEPA) introduced in 2022. This open source framework, paired with their cutting-edge world models, is sure to be a game changer in bringing the capabilities of world models further.
It’s not an easy path, but AMI Labs is on an ambitious journey. It intends to continue publishing influential research articles while fostering an innovative ecosystem that values the free exchange of ideas. This strategy offers a significant opportunity to foster collaboration among the many federal, academic, and industrial AI research institutions. Beyond that, it will make the field more productive as a whole.