{"id":23756,"date":"2024-11-25T10:29:22","date_gmt":"2024-11-25T02:29:22","guid":{"rendered":"https:\/\/www.1ai.net\/?p=23756"},"modified":"2024-11-25T10:29:22","modified_gmt":"2024-11-25T02:29:22","slug":"%e6%96%af%e5%9d%a6%e7%a6%8f%e5%a4%a7%e5%ad%a6%e3%80%81%e5%8d%8e%e7%9b%9b%e9%a1%bf%e5%a4%a7%e5%ad%a6%e7%a0%94%e7%a9%b6%ef%bc%9a1000-%e4%b8%aa-ai-%e6%99%ba%e8%83%bd%e4%bd%93%e9%a2%84%e6%b5%8b%e4%ba%ba","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/23756.html","title":{"rendered":"Stanford, UW Study: 1000 AI Intelligences Predict Human Behavior With Accuracy Up to 85%"},"content":{"rendered":"

As The Decoder reported today.Stanford University<\/a>,University of Washington<\/a>and Google DeepMind researchers have jointly developed an AI Agent<\/a>(AI) Agent<\/a>AI agents, also known as \"AI agents\", can realistically simulate human behavior in social experiments.<\/p>\n

\"724407d3j00snhiv8008nd000hs009cp\"<\/p>\n

The study noted that this type of simulation system could serve as a virtual laboratory thatHelp validate theories in economics, sociology, organizational and political science<\/strong>. The research team was able to accomplish this by conducting a study of\u00a0More than 1,000 representative Americans<\/strong>\uff08These AI intelligences were modeled by conducting interviews covering a wide range of ages, genders, educational backgrounds, and political affiliations.<\/p>\n

The system operates asCombining detailed interview transcripts with GPT-4o modeling<\/strong>. When a user asks a question to an intelligent body, the system loads the interview transcript andLet AI mimic respondents' answers<\/strong>. To generate these records, the research team administered each respondent'sTwo-hour-long interviews<\/strong>and transcribes the conversation into text using OpenAI's Whisper technology.<\/p>\n

The research team designed several tests to evaluate\u00a0AI's ability to predict human behavior<\/strong>, covering the General Social Survey, the Big Five personality assessment, and several behavioral economics experiments.<\/p>\n

The results show that AI based on interview data in theAccuracy of up to 85% in predicting social survey (GSS) questions<\/strong>, far beyond AI intelligences that rely solely on demographic information.<\/p>\n

The research team tested human performance against AI in five social science experiments, four of which showed thatAI results are highly consistent with humans<\/strong>The correlation coefficient is as high as 0.98. The correlation coefficient is as high as 0.98, suggesting that AI has an extremely strong ability to simulate human behavior.<\/p>\n

Interview-driven AI shows significant advantages over traditional demographic methods, especially in handling predictions of different racial groups and political views more accurately, while also effectively reducing bias between demographic categories.<\/p>\n

The research team has opened the dataset containing 1,000 AI models to GitHub with a two-tier access mechanism. Comprehensive data is fully open to researchers, while detailed data on individual participants is subject to special approval.<\/p>\n

Attach the address of the dataset:<\/span><\/p>\n