According to BiJie.com news, on June 20th (UTC+8), at the SuperAI Conference held in Singapore yesterday, Manus AI co-founder Tao Cheung pointed out that while current mainstream large models such as GPT-4 and Gemini already possess "intelligence at the doctoral level," they are still limited to "only writing and drawing" in reality, lacking actual execution capabilities. He said that this is precisely the reason for driving Manus's shift towards developing AI Agents—they can not only understand tasks but also proactively complete them. "What we need is not just a 'smart brain,' but a digital assistant that can truly complete tasks." The AI Agent trend is rapidly heating up in the Asian technology circle, with many companies transitioning from using language models such as ChatGPT to exploring AI Agent systems with execution capabilities. (Source: CNA)

#AI