This week, Sakana AI, a Tokyo-based startup, announced that it has appointed Jürgen Schmidhuber as its Chief Scientific Advisor. Schmidhuber, known as the "father of modern AI," is credited with foundational work in deep learning and world models, which have significantly influenced the current AI landscape.
Sakana AI aims to leverage Schmidhuber's expertise to steer its new RSI lab, which plans to develop a self-reinforcing research loop to enhance machine intelligence. The lab's mission is to gather a "critical mass of world-class experts" in Tokyo to advance its projects like the Darwin Gödel Machine and The AI Scientist.
Schmidhuber's work, particularly his 1987 thesis on meta-learning and recursive self-improvement, laid the groundwork for decades of research into how AI can continuously improve itself. Sakana AI is focusing on building autonomous systems that can understand, simulate, and interact with the physical universe, aiming to bridge Japan's legacy in neural network architectures and advanced robotics.
The appointment is targeted towards researchers and developers in the AI field, especially those interested in autonomous systems and physical AI. Sakana AI's initiatives are likely to attract attention from those seeking to innovate at the intersection of AI and robotics.
Work implications: Schmidhuber's involvement with Sakana AI could accelerate advancements in AI-driven autonomous systems, potentially impacting roles in robotics development and AI research.
Originally reported by The Decoder