If you've ever wondered how intelligent the artificial intelligence systems of major companies really are compared to humans, today you'll get a "standard" answer. A few days ago, a paper published on the Cornell University website caught the attention of the media. Titled *Intelligent Intelligence Evaluation and Intelligence Ranking Study*, it was authored by three Chinese scholars from the Chinese Academy of Sciences. The research was conducted at the Virtual Economy and Data Science Research Center, where they analyzed major AI systems such as Google, Microsoft Bing, Baidu, and Apple Siri, categorizing them into seven levels of intelligence.
The study aimed to create a unified benchmark for evaluating both human and artificial intelligence. To do this, researchers broke down the processes of "knowledge acquisition, mastery, innovation, and feedback" into 15 subcategories, including image and text recognition, common sense, calculation, translation, creation, and more. This multidimensional approach allowed for a more comprehensive assessment of AI capabilities.
In 2014, the team tested 50 AI systems across three age groups. Google, Baidu, and Sogou performed best, but they still lagged behind human intelligence. By 2016, the results showed significant progress in AI systems like Google and Baidu, though they were still far from matching the cognitive abilities of a 6-year-old child. Liu Feng, one of the researchers, noted that while AI has improved rapidly, creativity remains a challenge, and this category carries high weight in the evaluation.
The paper also proposed a seven-level classification system for AI, ranging from Level 0 (theoretical only) to Level 6 (capable of continuous innovation and knowledge generation). AlphaGo, for example, is currently classified as Level 3, as it lacks the networked collaboration required for Level 4.
The researchers believe that IQ testing for AI can help classify and guide the development of intelligent systems. They plan to conduct a third global AI IQ test in 2018, comparing AI progress with human cognitive development. Finally, Liu Feng clarified that the paper originated from the Chinese Academy of Sciences, not Cornell University, as some media had incorrectly reported.
For more smart TV and box information, visit Smart TV/Box Info Network Sofa Butler (http://), a leading platform offering insights, updates, and solutions for smart TVs and related technologies.
ETOP WIREHARNESS LIMITED , https://www.etopwireharness.com