India’s massive youth population—the much-touted demographic dividend—is set to come of age into an information environment rife with misinformation. In a 2025 American Political Science Review study Sumitra Badrinathan, Simon Chauchard, Florian Sichart, and I conducted across rural Bihar, more than half of the thousands of teenage students in our control group believed that exorcism can cure snakebites. Over 60 percent believed cow urine can cure COVID-19. The figures make it clear these were not fringe beliefs. Instead, they reflect how widely health misinformation has spread. While advances in technology have made information access quicker and easier than ever, the spread of false information through trusted family WhatsApp groups, social media platforms, and word-of-mouth networks has been linked to several undesirable societal outcomes. From unscientific health remedies to fabricated political conspiracies, misinformation about health, politics, and minorities shapes young people’s attitudes about vaccines, elections, and out-groups. Such attitudes can even influence critical behaviors—such as whether to seek professional medical help—with important implications for individual and community health. Can education offer a solution? Our study suggests that classroom interventions can work, even in the most challenging settings.In response to the threat posed by misinformation, several governments around the world, including Finland and New Jersey, now mandate media literacy in schools. While international organizations and forums have been advocating such educational interventions, adoption remains very limited. According to a UNESCO report among 194 member states, fewer than 50 percent have formally integrated these initiatives into their education systems in any capacity, and adoption is much lower among Asian and African countries. The report highlights several main barriers to implementation, including a lack of dedicated funding, inadequate teacher training, and the absence of a coordinated policy framework. Before governments invest into media literacy programs, there is a need for systematic empirical evidence on whether such interventions build citizens’ capacity to navigate complex information environments and counter misinformation.My co-authors and I attempted to test the efficacy of such interventions in rural Bihar with large-scale experimental evidence of classroom-based media literacy education. Bihar presents a unique and challenging context for information and media literacy interventions. The state has one of the highest school dropout rates in India. There is a strong cultural attachment to non-scientific medical approaches, like folk remedies, which can conflict with evidence-based medical approaches when treating serious conditions, as we discovered in our baseline surveys. There is also low diffusion of internet access, with only 19% of our sample reporting internet usage, making students particularly vulnerable to misinformation when they do encounter online content. False health information circulates widely through social networks, often presented with the same authority as scientifically validated information. Hence, insights from Bihar – a potentially hard case for the success of such interventions—may provide valuable lessons for policymakers and other stakeholders across the country.BIMLI Our study, the Bihar Information and Media Literacy Initiative (BIMLI), involved carrying out a randomised controlled trial encompassing over 13,500 students in grades 8 through 12 across 583 villages in Bihar. My co-authors and I designed and implemented four 90-minute “literacy” sessions over roughly 14 weeks in 100 government-run community libraries across 32 districts. Around each library, nearby villages were selected and randomly assigned to either the treatment group or the control group. From each village, classrooms were created with 20 to 24 students. The treatment villages received a media and information literacy course, while the control villages received basic English-language instruction, allowing us to hold constant classroom participation and engagement while varying only the lecture content. This initiative was delivered as an official government course through our partnership with the Bihar Rural Livelihoods Promotion Society (commonly known as Jeevika), which significantly boosted legitimacy and enrollment rates.Figure 1: A trained educator provides a 90-minute health information training session.The curriculum and training of educators (who delivered the program), which was co-designed and implemented with DataLeads, focused on developing critical evaluation skills for health information sources, teaching students to identify misleading or manipulative health content, providing practical tools for verifying information credibility, and building capabilities for responsible information sharing within communities. The program included specific modules on assessing the reliability of different types of health advice, understanding the difference between anecdotal and scientific evidence, and recognising common tactics used to make false information appear credible.Discerning Information Credibility BIMLI had significant positive effects across multiple outcomes. Students who received the media literacy course showed a 35 percent improvement in accuracy in discerning true from false health headlines, which were locally salient. This improvement was not simply about memorizing specific facts, but rather about developing systematic approaches to evaluating information credibility. Source discernment improved by 42 percent, indicating that students developed a sophisticated understanding of how to assess the reliability of different information sources. Perhaps most importantly for public health outcomes, students were 27 percent less likely to forward questionable health claims and more likely to verify information before sharing it with family members or friends. This change in sharing intention has the potential to slow the spread of health misinformation in community networks. The intervention also achieved a 15 percent reduction in preference for non-scientific medicine for serious health conditions, primarily by reducing vaccine hesitancy and stated reliance on alternative forms of medicine for serious illness. This finding is significant because research indicates that beliefs related to group identity are highly resistant, and alternative medicines, which have existed for generations, are closely tied to these identities. Figure 2: Treatment effects on key outcomes. Points show improvements measured in standard deviations (a statistical measure of effect size). Bars show the range of uncertainty.Our study also found evidence of lasting impact, with effects remaining significant in the four-month follow-up across most primary outcome measures. Moreover, the intervention also produced significant spillover effects to parents and family members who did not directly participate in the program. Even the parents of participating students showed improved information-evaluation abilities, providing evidence of “trickle-up” education from children to adults. These findings challenge conventional assumptions about information flow within families and suggest that well-designed youth programs can have broader community impact. In addition, the skills appeared to transfer beyond health misinformation (the curriculum’s focus) to political rumor. Students who received the media literacy sessions were better at identifying true versus false political news. These findings suggest that inoculating against misinformation in one domain (in our case, health misinformation) can also build capacity against other misinformation (in this case, political misinformation). Despite the encouraging findings, a limitation of our design is that it does not allow us to systematically attribute the results to the different elements of the intervention—curriculum, well-trained facilitators, or partnership with a trusted government agency, peer interaction among students; future research should aim on isolating these to get more specific insights for policy.Policy counterpointTwo lessons stand out for policymakers considering classroom-based media literacy. First, the intervention worked in a setting characterised by low literacy, limited digital access, and deeply rooted identity-based misinformation. That the initiative generated large and persistent effects under these constraints underscores the power of repeated, interactive, offline instruction delivered by trusted local institutions. Second, the acquired skills are transferable. Students did not merely memorise health facts; they learned how to evaluate information more broadly, applying those skills even to political news that was never discussed in class. This kind of transfer is critical for policy relevance: governments need programs that prepare students for a wide array of evolving misinformation threats.These findings also offer a counterpoint to more restrictive approaches. Australia recently enacted legislation banning children under 16 from social media platforms, reflecting growing concern about online harms. While such measures stem from legitimate worries, outright bans are difficult to enforce and do little to equip young people with the skills they will eventually need to navigate digital spaces. Education-based interventions offer an alternative: by building critical evaluation skills early, they foster resilience rather than dependence on restrictions. If children can learn to assess content critically before they encounter it unsupervised, the case for blanket bans weakens considerably. Investing in media literacy may ultimately prove more sustainable – and more respectful of individual agency—than attempting to wall off the digital world entirely.In Bihar and elsewhere, misinformation has fuelled health risks, polarisation, and even mob violence. Rumours about vaccines, religious communities, and public health measures have spread rapidly, often with devastating consequences. In the coming years, policymakers should actively seek more evidence to determine whether this study’s success in Bihar can be replicated in other Indian states, whether the positive effects last beyond four months, and how curricula can adapt to counter AI-generated misinformation. Achieving this will require ongoing collaboration among educators, technologists, and researchers to monitor the evolving information landscape. A media literacy program that is designed once and never revisited risks becoming outdated, leaving students unprepared for a rapidly changing world. India’s demographic dividend could become a misinformation liability – or, with the right media literacy efforts and investments in education, its greatest asset in building a more discerning citizenry.Priyadarshi Amar is a Postdoctoral Researcher at the University Carlos III (Madrid) and Instituto Juan Linz.This article originally appeared on the website of the University of Pennsylvania’s Centre for Advanced Study of India, as part of the ‘India in Transition’ series. It has been republished with permission.