Demystify Artificial Intelligence and Technology Outlook

The term “Artificial Intelligence (AI)” was coined in 1956. The field evolved slowly in the first 50 years. Yet, we have witnessed rapid advances in AI in the last decade. One example is the availability of the ChatGPT service. Questions have been raised about the role of AI in human society and civilization, e.g., whether AI will replace human intelligence (HI), how HI and AI complement each other, etc. In this talk, I will shed light on two issues. First, it is common to divide AI into two categories: strong AI and weak AI. Strong AI is close to HI while today’s AI still belongs to the weak AI category. There is a substantial gap between weak and strong AI.  AI should be human’s servants and tools (rather than a master) from such a perspective. Second, I will point out the limitations of today’s AI tools based on deep learning (DL). DL is neither interpretable nor sustainable. It is critical to develop an alternative AI methodology that is interpretable and sustainable. I have conducted research on this problem since 2015 and propose a green learning (GL) paradigm. GL significantly reduces the model size and the complexity of DL models, yielding green AI solutions. The GL methodology will be briefly presented.