热点问答丨美消费电子展上演讲嘉宾如何谈AI

Group 1: AI Development Trends - The 2026 CES highlighted the evolution of AI from cloud-based virtual systems to physical applications in real-world scenarios [1] - AI's next phase includes breakthroughs in computing power, with a need to increase global computing capacity by 100 times in the coming years to support widespread AI adoption [2] - The concept of "Physical AI" is emerging, focusing on AI systems that not only understand but also interact with the physical world, with applications in autonomous driving and industrial automation [3] Group 2: Computing Power and Infrastructure - The current global AI user base exceeds 1 billion and is expected to surpass 5 billion, necessitating significant advancements in computing capabilities [2] - Companies like NVIDIA are developing integrated AI computing platforms, such as the "Vera Rubin" system, which includes six chips designed to enhance model training efficiency and reduce inference costs [2] - The industry is shifting towards a full-stack collaborative design approach, integrating chips, networks, and storage to create comprehensive AI solutions [2] Group 3: Intelligent Agents and Personalization - The year is anticipated to be significant for "proactive intelligent agents," which can autonomously perform tasks and understand complex goals, moving beyond traditional reactive AI [4] - Lenovo introduced its first personal super intelligent agent, Lenovo Qira, which connects various devices and supports context awareness and user preference prediction [5] - The integration of personal, enterprise, and public intelligence through hybrid AI is seen as a key pathway for personalized and diverse AI development [5]