Each week, a new tech headline boldly proclaims that “Y will kill X”. This dynamic duo of Y and X has seen countless iterations throughout history. From photography’s supposed demise of painting to MOOCs’ (massive open online course) alleged overthrow of universities, and even the literal “video killed the radio star,” the narrative of technological disruption is a familiar tune. Now, the spotlight shines on ChatGPT’s Study Mode, heralded as the impending executioner of India’s private-tuition industry.
That prediction is not entirely baseless. In cities such as Delhi, hiring a personal tutor typically can cost over Rs 1,000 per hour, with rates climbing for senior-secondary science and competitive-exam coaching. Households pay these fees because tutors offer three things: Individualised explanations, structured practice and instant doubt-clearing. OpenAI’s newly released Study Mode positions itself as a free, always-available provider of exactly those services.
OpenAI describes Study Mode as “a learning experience that helps you work through problems step by step instead of just getting an answer.” The system begins with a short diagnostic exchange that gauges prior knowledge, then leads learners through Socratic prompts, targeted hints, and periodic knowledge checks. Explanations are delivered in sequential, bite-sized sections that highlight links between ideas — an approach the company says was shaped with teachers, scientists, and learning-science researchers to manage cognitive load and promote metacognition. Students can toggle the coach on or off at any time, retaining the option of a conventional Q-and-A when speed is essential.
OpenAI says this is only a first step. Upcoming iterations will add clearer visualisations for dense or text-heavy concepts, goal-setting and progress-tracking tools that span multiple chats, and even deeper personalisation tuned to each learner’s long-term objectives. These upgrades are expected to arrive over the next product cycles and will likely benefit from the higher reasoning capacity and multimodal output that industry observers anticipate in the newly launched GPT-5 model.
In practice, these design choices prove effective. During trials with Grade 10 algebra problems and first-year university economics questions, Study Mode consistently withheld answers until a partial line of reasoning was offered, adjusted the depth of explanation after each learner’s response, and revisited earlier misconceptions in later prompts. The immediate, structured feedback felt comparable to a diligent human tutor, especially when the model summarised progress at the end of a session and suggested next steps. Although occasional inaccuracies still surfaced, the overall experience was sufficiently clear and adaptive to leave a strong impression of genuine, learner-centred guidance. However, Study Mode’s effectiveness varies, and it generally excels in STEM and English. While free access may be limited, a subscription typically offers seamless use.
From a purely functional standpoint, Study Mode matches or exceeds what many tutors offer. It is online and available 24×7 and supports multiple languages. Its cost advantage is pronounced; even if a family opts for ChatGPT Plus to access the fastest model, the subscription is roughly $20 a month, far below a premium metro-city tuition cost. For learners who live in districts where qualified tutors are scarce or unaffordable, the prospect of an AI coach on a smartphone is undeniably attractive.
Yet the promise of an on-demand AI tutor will matter only if learners can reach it. In many rural households, a single handset is shared among several family members, mobile data is rationed, and electricity is still unreliable. Under such conditions, a text-heavy AI tutor that assumes stable connectivity risks reinforcing the very disparities it is meant to reduce. Unless governments, telecom providers and ed-tech firms work together towards these gaps, the students who can already afford private tuition will also be the first to benefit from AI guidance, doubling their advantage rather than closing the gap.
Pedagogically, human tutors also retain several advantages. They notice when a pupil’s eyes glaze over, adjust pace on the fly and can provide the social accountability many adolescents need to keep learning. Large language models (LLMs) continue to produce occasional hallucinations; a vigilant tutor can catch a faulty hint delivered by the AI before it hardens into a misconception in the student’s mind. Human tutors remain indispensable for providing social-emotional support, addressing special needs, and conducting real-time diagnostic questioning.
What appears more likely than complete displacement is a clear realignment of responsibilities. Private tutors may spend less time working out every trigonometric identity in a notebook and more time acting as meta-coaches, guides who teach students how to frame effective questions for the AI, interpret the model’s hints, and cross-check its explanations. In this new arrangement, a tutor’s primary value lies in curating trustworthy resources, diagnosing subtle misconceptions that an algorithm might miss, and designing structured study plans that weave human interaction with AI-generated practice tasks.
The human element of mentorship becomes even more critical in a tech-enabled environment. Many students need someone who notices when motivation dips, encourages reflection on learning strategies, and contextualises knowledge within future goals and career aspirations. A mentor can guide learners in developing habits such as setting objectives, monitoring progress, and adjusting approaches when difficulties arise. For first-generation learners who lack proximity to subject experts, this blended approach is particularly powerful: an AI tutor provides on-demand academic guidance, while a trusted adult or community volunteer nearby provides the encouragement, accountability and context that machines cannot. Far from rendering tutors obsolete, Study Mode raises the bar, asking them to become architects of learning environments where human insight and machine efficiency work in tandem.
So will ChatGPT’s Study Mode kill home tuitions? Probably not, at least, not in the melodramatic sense of sudden extinction. What it will do is raise expectations. Any tutor who merely recites answers from last year’s guidebook will face tough competition from a free AI that never tires. Educators who reinvent themselves as mentors in critical thinking, metacognition and responsible AI use are likely to become more essential, not less.
The classroom of the near future will rest on a tripod: An inquisitive student, a capable AI assistant and a perceptive human guide. Remove any leg, and the learning process wobbles. Study Mode may narrow educational inequities where connectivity exists, but it also challenges policymakers and technology companies to extend infrastructure and craft clear guidelines. The technology has set a higher bar; it is up to the educators to decide whether that bar becomes a stepping-stone or a stumbling block.
The writer is an education consultant specialising in AI for educators, curriculum design, teacher training, ed-tech strategy, and innovative math-learning initiatives
