In a recent trip to Beijing, Nvidia's founder and CEO, Jensen Huang, shared insightful career advice for young graduates navigating the rapidly evolving field of artificial intelligence. Reflecting on his own path, Huang—who earned an electrical engineering degree from Oregon State University in 1984 and a master's from Stanford in 1992—stated that if he were a 20- or 22-year-old student graduating in 2025 with the same drive, he would shift his focus from software and coding to the physical sciences.
Huang emphasized that the next era of AI demands a deep grasp of real-world principles, such as the laws of physics, friction, inertia, and cause-and-effect relationships. He explained that AI has progressed through distinct phases: starting with "Perception AI" around 12-14 years ago, highlighted by the 2012 breakthrough of AlexNet in computer vision; moving into "Generative AI," where systems can interpret and create content across formats like text, images, and code; and now entering "Reasoning AI," enabling machines to solve novel problems and function as adaptive digital agents, as pursued by companies like Microsoft and Salesforce.
Looking ahead, Huang predicted the rise of "Physical AI," which will empower machines to interact meaningfully with the physical environment. This includes developing skills like object permanence (recognizing that objects exist even when out of sight), predicting forces (such as how much grip is needed to hold an item), and situational awareness (inferring unseen elements, like a pedestrian behind a car). When applied to robotics, Physical AI could revolutionize manufacturing by creating highly automated factories in the U.S. to combat global labor shortages over the next decade.
This guidance echoes sentiments from other tech leaders, including Telegram's Pavel Durov, who advocates for mathematics, and Elon Musk, who promotes physics paired with math, underscoring a return to foundational sciences for logical reasoning and innovation in a robotics-driven world.economictimes.indiatimes.com Huang's message serves as a call to action for students: mastering the mechanics of the real world, rather than just data manipulation, will be key to unlocking AI's transformative potential in areas like robotics and beyond.
IMG SRC: https://upload.wikimedia.org/wikipedia/commons/3/36/Jensen_Huang_20231109_%28cropped2%29.jpg
Laisser un commentaire
Les commentaires sont approuvés avant leur publication.
Mochy Group
Auteur