Google’s chief AI scientist, James Manyika, emphasizes that artificial intelligence systems require multiple generations of development to achieve human-like learning capabilities. In his discussion at a conference in Montreal, Manyika highlighted that current AI systems, despite their impressive capabilities, still lack the fundamental ability to learn as efficiently as humans do. He points out that while humans can learn new concepts from just a few examples, AI systems typically need massive amounts of data and training. The scientist stressed that developing AI systems that can learn more efficiently is a crucial challenge that will likely take several generations to solve. Manyika also addressed concerns about AI’s impact on jobs, suggesting that while AI will transform many roles, it’s more likely to augment human work rather than completely replace workers. He emphasized the importance of responsible AI development and the need for careful consideration of its societal implications. The discussion touched on Google’s approach to AI development, which focuses on making systems more efficient and environmentally sustainable. Manyika concluded by highlighting the importance of continued research in areas like few-shot learning and transfer learning, which could help AI systems become more adaptable and efficient learners. The overall message underscores the long-term nature of AI development and the need for patience in achieving more sophisticated AI capabilities.