Princeton Professor: Students Face a Choice Regarding the Use of AI
Arvind Narayanan, a professor of computer science at Princeton University, said that students face a difficult choice: they need to master artificial intelligence tools for their future careers, but overreliance on them could hinder the development of basic skills. He shared these thoughts with Fortune.
Narayanan, co-author of the book *AI Snake Oil*, does not consider generative AI to be a useless technology. In his view, knowledge workers can already use such tools for research, testing hypotheses, data analysis, and software development. At the same time, the professor is critical of systems that are claimed to be capable of reliably predicting human behavior or future outcomes—particularly when it comes to decisions regarding hiring, insurance coverage, bail, or police activities.
According to the researcher, resistance to AI stems from several different concerns: fear of job loss, distrust of large tech companies, anxiety about the influence of billionaires, environmental costs, and the social consequences of the technology. Young people are also concerned about which skills they need to maintain in a labor market where AI is becoming increasingly widespread.
Read more breaking news on the UA.News Telegram channel.
Narayanan believes that automation does not necessarily mean direct job cuts. The use of AI may increase the need for review, oversight, and verification of its results. The risk, he says, lies in the fact that employees may be left with formal responsibility for the decisions of automated systems over which they have insufficient influence or control.
The professor also highlighted the differences between professions. Developers can interact with AI while working, review the generated code, and maintain control over the process. For artists, a request to the system can immediately generate a finished image, bypassing a significant part of the creative process. According to Narayanan, the beneficial use of AI should help users deepen their understanding and retain the ability to independently evaluate the system’s results.
