A research team led by Associate Professor Ye Weiming from Peking University HSBC Business School and Li Qian, a faculty associate at Arizona State University's Walter Cronkite School of Journalism and Mass Communication, explored how the public discusses AI-assisted learning. Their findings, published in the journal Comunicar, examine the cognitive, behavioral, and ethical effects of using artificial intelligence to help people learn, understand, and apply knowledge. AI has made it easier to access information and complete complex tasks more efficiently. However, as its use grows, so do concerns about inaccurate information, reduced independent thinking, issues with plagiarism, and unclear lines of responsibility. The researchers analyzed AI-assisted learning through three dimensions—cognition (how people think), behavior (how people act), and ethics (moral considerations)—and three stages of knowledge construction: acquisition (learning new information), comprehension (understanding it), and application (using it in practice). To gather insights, they studied discussions on Reddit from 2013 to 2025. After collecting 20,174 texts, they narrowed it down to 10,503 valid posts following data cleaning. Using a method called the Biterm Topic Model (BTM) and qualitative analysis, the team identified key public concerns about how AI is transforming learning and knowledge building. During the knowledge acquisition phase, users praised AI for its speed and ability to organize information. However, they also expressed anxiety over factual errors and flawed reasoning, which made them question the reliability of AI-generated content. In the comprehension phase, AI helped simplify complex ideas, but overreliance on it raised concerns about reduced critical thinking and independent reasoning. Some users responded by limiting their use of AI and taking steps to reconstruct knowledge on their own. During the application phase, AI offered new ways to solve problems and reduced the costs of trial and error. But it also sparked worries about plagiarism, intellectual property rights, fairness, and who is ultimately responsible for mistakes. Public attitudes toward AI-assisted learning followed a cycle of initial trust, growing doubt, and eventual recalibration. AI use evolved from individual experimentation to more collaborative approaches, including comparing results, verifying information together, and checking facts within communities. The study suggests that AI is no longer just a tool but is increasingly seen as an intellectual partner, a behavioral facilitator, and an ethical discussion partner. The researchers argue that AI-assisted learning involves an ongoing negotiation between efficiency and trust, autonomy and responsibility. They recommend a governance approach that is transparent, inclusive, and adaptable to keep pace with the evolving role of AI in education.