Sridhar Vembu has raised alarms regarding the increasing reliance on AI tools among high school students, particularly in the context of competitive examinations. He expressed these concerns after reviewing a study that examined the effects of AI-assisted learning on student performance. According to Vembu, while AI may seem advantageous in the short term, it could undermine a student’s ability to think independently.

His remarks followed a viral post by Rimsha Bhardwaj on X, referencing a 2025 research paper published in the Proceedings of the National Academy of Sciences (PNAS). Vembu’s critique highlights the potential long-term consequences of over-reliance on AI technologies in educational settings.

Findings from the Research Study

The study conducted by Hamsa Bastani and her colleagues at the Wharton School involved nearly 1,000 high school students in Turkey. The students were divided into three groups during mathematics practice sessions. One group utilised a traditional GPT-4 chatbot that provided direct answers, another group used a tutor-style version that offered hints instead of complete solutions, and the third group studied without any AI assistance.

Initially, the AI-assisted groups displayed significantly higher performance levels. Students using the standard GPT model reportedly solved 48 per cent more questions, while the tutor-style AI group completed 127 per cent more compared to those who practiced without AI. However, the results of a subsequent test, taken without access to AI tools, revealed a different picture. The group that relied on the standard GPT interface scored 17 per cent lower than those who had studied independently.

Researchers noted a concerning trend: many students had started depending on the chatbot for problem-solving rather than grasping the underlying mathematical concepts. This dependency created a false sense of understanding, and students who relied heavily on AI often displayed greater confidence in their abilities, which did not align with their actual performance in examinations.

The Broader Implications of AI in Education

Despite growing apprehensions regarding overdependence, AI tools are becoming an integral part of learning and work environments. Major technology firms such as Meta and Google are encouraging their employees to incorporate AI into tasks related to coding, research, and productivity.

This trend has extended into education; for instance, Google recently integrated mock test functionalities into its Gemini AI platform for JEE Main aspirants. Students can generate practice exams by requesting prompts, with the AI creating tests from question banks sourced from well-known educational platforms. Following the tests, the AI analyses student performance, identifies weaknesses, and explains answers, effectively shaping a new study approach.

Vembu has also touched on aspects of human life that technology may not displace, emphasising roles driven by purpose rather than monetary gain, such as teaching or caring for others. He asserts that as automation evolves, society may progressively shift towards these purpose-led activities, which could remain insulated from AI’s reach.

In the broader dialogue on technology’s impact, Vembu posits that the critical issue extends beyond economic gain. He envisions a future where AI enhances productivity, resulting in an abundance of goods and services. However, he raises concerns regarding how society will manage and distribute this surplus, particularly given potential shifts in employment opportunities and income stability.

Long or Short, get news the way you like. No ads. No redirections. Download Newspin and Stay Alert, The CSR Journal Mobile app, for fast, crisp, clean updates!

App Store –  https://apps.apple.com/in/app/newspin/id6746449540 

Google Play Store – https://play.google.com/store/apps/details?id=com.inventifweb.newspin&pcampaignid=web_share