An AI lab backed by Sequoia Capital and other top investors is taking a bold new approach to artificial intelligence, one that challenges the dominant paradigm of training models on massive internet datasets. Flapping Airplanes, a newly launched startup founded by brothers Ben and Asher Spector along with co-founder Aidan Smith, has secured $180 million in seed funding from a roster of heavyweight investors including Google Ventures, Sequoia, and Index Ventures. Unlike most AI labs that focus on scaling up data and compute to build ever-larger models, Flapping Airplanes is pursuing a different path: creating AI systems that learn more like humans—efficiently, with far less data, and with a focus on understanding and creativity. The team believes the human brain represents not the limit of what’s possible, but the starting point. The founders argue that current AI models, while powerful, are fundamentally inefficient. They require enormous amounts of data and energy to train, often mimicking patterns without true comprehension. Flapping Airplanes aims to reverse that trend by developing architectures and training methods that prioritize data efficiency and cognitive-like learning. In a recent episode of TechCrunch’s Equity podcast, AI editor Russell Brandon spoke with the founding trio about why such a large investment was made for a company with no product yet. The answer lies in a growing belief among investors and researchers that the era of simply scaling up models may be reaching its limits. The next breakthrough, they argue, will come not from more data, but from smarter ways of learning. The team is placing a strong emphasis on creativity and original thinking over traditional credentials. They’re assembling a team that values curiosity and unconventional problem-solving, reflecting their mission to build AI that doesn’t just replicate human knowledge, but extends it. The $180 million funding will be used to hire top talent, build experimental infrastructure, and run large-scale research initiatives focused on efficient learning. The founders believe that once AI can learn with human-like efficiency, it could unlock entirely new capabilities—such as rapid adaptation to new tasks, deeper reasoning, and the ability to generate truly novel ideas. While many labs have quietly abandoned the pursuit of data-efficient AI in favor of the proven path of scale, Flapping Airplanes is betting that the future belongs to systems that think differently.