谷歌DeepMind的首席执行官Demis Hassabis在近期接受《连线》(WIRED)杂志采访时表达了对人工智能（AI）未来发展持乐观但谨慎的态度。尽管OpenAI的ChatGPT在科技界引发了广泛的关注，Hassabis认为，AI领域最重大的突破仍未到来，而这并不仅仅依赖于芯片技术的提升。
Hassabis强调，要实现通用人工智能（AGI）的愿景，需要的不仅仅是计算能力的极致扩展。他认为，创新是推动AI前进的关键，特别是在代理学习、强化学习和规划等领域。他提到，DeepMind早在AlphaGo时代就开始在这三个方面进行深入研究，这些工作为AI技术的进步奠定了坚实的基础。
Hassabis的言论反映出，尽管当前AI技术已经取得了显著的进步，但要达到人类智能的广度和深度，还需要在算法、理解和决策能力上实现更多的突破。这表明，AI研究的未来将更加注重理论与实践的结合，以及如何让机器更好地理解和适应复杂环境，而不仅仅是提升硬件性能。随着全球科技巨头如DeepMind等在AI领域的持续探索，我们可以期待更多创新性的成果将逐渐显现。
英语如下：
News Title: “DeepMind CEO: Major AI Breakthroughs Lie Ahead Beyond Chip Advances”
Keywords: DeepMind, AI breakthroughs, computational innovation
News Content: Google’s DeepMind CEO, Demis Hassabis, expressed a cautious yet optimistic outlook on the future of artificial intelligence (AI) in a recent interview with Wired magazine. Despite the widespread attention generated by OpenAI’s ChatGPT, Hassabis believes that the field’s most significant breakthroughs are yet to come, not solely dependent on advancements in chip technology.
Hassabis emphasized that realizing the vision of artificial general intelligence (AGI) requires more than just pushing the boundaries of computational power. He argues that innovation is crucial to propel AI forward, particularly in areas such as agent learning, reinforcement learning, and planning. He noted that DeepMind has been conducting in-depth research in these three domains since the AlphaGo era, laying a solid foundation for the progress of AI technologies.
Hassabis’s remarks reflect that while AI has made remarkable strides, achieving the breadth and depth of human intelligence necessitates further breakthroughs in algorithms, understanding, and decision-making capabilities. This suggests that the future of AI research will focus more on the integration of theory and practice, and enhancing machines’ ability to understand and adapt to complex environments, not just improving hardware performance. With ongoing exploration in the AI domain by global tech giants like DeepMind, we can anticipate more innovative成果 to emerge gradually.
【来源】https://www.wired.com/story/deepmind-ceo-demis-hassabis-interview-artificial-intelligence-scale/
Views: 1