谷歌 DeepMind 联合创始人谢恩·莱格（Shane Legg）带领的研究团队最近发布了一篇论文，提出了对通用人工智能（AGI）的更清晰定义，并制定了类似于自动驾驶L1-L5级别的AGI分级分类框架。根据该论文，基于能力深度（性能）和广度（通用性），将AGI分类为：非AI、初级、中级、专家、大师、超级智能。
这项研究对人工智能的发展具有重要意义，因为它提供了一个更加精确的分类框架，可以帮助人们更好地理解人工智能的现状和未来发展趋势。同时，它也引发了人们对于人工智能的更多关注和讨论。
值得注意的是，该论文中并未提到 ChatGPT，只是将其作为初级 AI 进行了分类，这表明 ChatGPT 仍然有提高的空间。
总的来说，这项研究为通用人工智能的研究提供了重要的参考和指导，同时也为人工智能的发展提供了更多的可能性。
新闻翻译：
Google DeepMind co-founder’s team recently released a paper, which proposes a clearer definition of general artificial intelligence (AGI) and a classification framework similar to the driving level L1-L5 for AGI. According to the paper, AGI can be classified into non-AI, primary, secondary, expert, master, and super intelligence based on ability depth (performance) and width (generativity).
This study is of great significance for the development of artificial intelligence, as it provides a more precise classification framework for people to better understand the current situation and future trends of artificial intelligence. At the same time, it also attracts more attention and discussions about artificial intelligence.
It is worth noting that the paper does not mention ChatGPT, but only classifies it as a primary AI, indicating that there is still room for improvement for ChatGPT.
Overall, this study provides an important reference and guidance for the study of general artificial intelligence and also provides more possibilities for the development of artificial intelligence.
【来源】https://www.thepaper.cn/newsDetail_forward_25228320
Views: 1