top of page

By:

Anil D. Salve

21 March 2026 at 8:11:09 pm

A Drought of Wisdom in an Ocean of Information

We live in an age where information is everywhere. With a single click, we can access news from across the world, thousands of opinions, countless videos and endless claims. Knowledge has never been so accessible. Yet, in this overwhelming flow of information, something increasingly scarce is discernment-the ability to distinguish what deserves our belief from what deserves our doubt. Information and knowledge are not the same. Reading, watching or hearing something does not automatically...

A Drought of Wisdom in an Ocean of Information

We live in an age where information is everywhere. With a single click, we can access news from across the world, thousands of opinions, countless videos and endless claims. Knowledge has never been so accessible. Yet, in this overwhelming flow of information, something increasingly scarce is discernment-the ability to distinguish what deserves our belief from what deserves our doubt. Information and knowledge are not the same. Reading, watching or hearing something does not automatically make it true. We need the ability to verify information, understand its context, recognise its purpose and only then form an opinion. In the digital age, this ability has become one of the most important forms of wisdom. Social media has transformed the way information travels. WhatsApp, Instagram, Facebook and other platforms have made communication faster, wider and more accessible. But the same speed that connects society can also overwhelm it. Information often reaches us before facts do. A headline is read before the full story. A short video is watched without knowing what happened before or after it. A photograph is circulated without its original context. A message arrives with the familiar words, “Forwarded as received,” and within minutes it may reach thousands of people. By the time the truth catches up, the rumour may already have done its work. The problem becomes even more serious when misinformation is deliberately created and circulated. Rumours may be planted not merely to mislead individuals but to divert public attention from an important issue, create suspicion, damage someone's reputation or push society towards a particular narrative. A distraction does not always need to be completely false. Sometimes a small piece of truth is mixed with exaggeration, selective facts or emotional language. The result can be more powerful than an outright lie because it appears believable. This is how narratives are manufactured. Social media platforms can become powerful tools in this process. A repeated claim begins to look familiar, and familiarity is often mistaken for truth. A provocative post attracts attention. Thousands of reactions create the impression that “everyone” believes it. One emotional video triggers another, comments intensify the mood, and before facts have an opportunity to speak, public sentiment may have already shifted. This is where mob psychology enters the picture. A society can move from calm to anger, suspicion to trust, or optimism to fear with surprising speed. People who would ordinarily think carefully may react differently when they see thousands of others expressing outrage, fear or hostility. The individual begins to follow the crowd, and the crowd begins to reinforce the individual. The danger is not limited to negative emotions. The same mechanism can also be used positively-to mobilise people for social causes, humanitarian assistance, disaster relief, public awareness and community action. Technology itself is neither wise nor foolish. Its impact depends on how information is created, amplified and consumed. The greatest danger, therefore, is not that we have too much information. It is that we may lose the habit of thinking while consuming it. We are increasingly becoming consumers of information rather than creators of thought. Everyone has access to information today, but access does not necessarily mean understanding. The boundaries between a rumour and a report, an opinion and a fact, popularity and truth are becoming increasingly blurred. A post does not become true because thousands of people have liked it. A video does not become evidence merely because it has gone viral. An opinion does not become a fact because it is expressed confidently. An idea does not become correct simply because it confirms what we already believe. This last point deserves particular attention. We often accept information not because it is accurate, but because it is comfortable. We are naturally attracted to information that confirms our existing beliefs and suspicious of information that challenges them. In the digital world, this tendency can be amplified by algorithms that repeatedly expose us to content similar to what we have already watched, liked or shared. Gradually, we may find ourselves living inside an information bubble where we hear many voices-but mostly voices that agree with us. That is not diversity of information. It is an illusion of certainty. Discernment does not mean doubting everything. It means thinking before believing. It means asking simple but powerful questions: Who is saying this? What is the evidence? What is the context? What may be missing? Who benefits if I believe it? And have I heard the other side? These questions may take only a few seconds, but they can prevent hours, days or even years of misunderstanding. This responsibility is particularly important for the younger generation. Children today are growing up in an environment of unprecedented information exposure. They can encounter more information in a few hours than earlier generations might have encountered in weeks. Therefore, education cannot merely teach students how to collect, remember and reproduce information. It must teach them how to question, verify, analyse and interpret it. A student who can write the correct answer in an examination may be academically successful. But a student who knows how to ask the right question is better prepared for life. The classroom of the future must therefore produce not merely informed students, but thinking citizens. They must learn that forwarding a message is also a form of responsibility. Sharing a post is not an innocent act when it can affect someone's reputation, create public fear or influence social behaviour. The challenge becomes even greater in the age of Artificial Intelligence. AI can generate information, images, audio and video with extraordinary speed and sophistication. The distinction between what is real and what merely appears real may become increasingly difficult. In such a world, human judgment becomes more valuable, not less. Technology can produce an answer in seconds. It cannot automatically tell us whether that answer deserves our trust. Some answers require experience. Some require context. Many require the ability to think independently. We have more information, more technology and more connectivity than any previous generation. Yet our real challenge is to preserve something technology cannot manufacture for us: the capacity to think for ourselves. The ocean of information will only become larger. The question will no longer be, “How much information do we have?” The more important questions will be: What should we accept? What should we question? What should we verify? And what should we reject? No technology can completely answer these questions for us. No number of likes can substitute for evidence. No viral trend can replace judgment. No majority can turn misinformation into truth. That responsibility belongs to human beings. Information can tell us what happened. Discernment helps us understand what it means, why it happened, what may be missing from the story and how we should respond. Perhaps, therefore, the greatest educational challenge of our time is not to teach people how to find more information. They already know how to do that. The greater challenge is to teach them how to pause-to pause before believing, to pause before reacting, to pause before forwarding and to pause before joining the crowd. Because sometimes the most intelligent response to information is not an immediate reaction, but a moment of reflection. The ocean is already around us. It is growing deeper every day. What we need now is not more information to swim through it-but the wisdom to know which direction to take. (The writer is the Principal of Podar International School, Ausa, Latur. Views personal.)

Bridging the Epistemic Gap: Why India Needs Ethical AI Audits

Writer: Sagari Gupta
Sagari Gupta
May 18
5 min read

Without measuring contextual failure, India’s AI governance risks mistaking computational sophistication for genuine public utility.

India’s AI governance framework has a measurement problem. The India AI Mission has onboarded more than 38,000 GPUs, allocated Rs 10,300 crore for AI infrastructure, and built AIKosh into a national dataset repository holding 7,541 datasets across 20 sectors. MeitY’s India AI Governance Guidelines, released on 5 November 2025, commit the government to fairness, equity, and human-centric design as core principles. What neither the infrastructure nor the guidelines measure is whether AI systems actually work for the populations they are meant to serve.


Accuracy sans Adequacy

India’s audit regime evaluates whether AI models are technically accurate. It does not evaluate whether they are informationally adequate for the users who receive their outputs. The two are not the same, and in India’s context, the gap between them is wide.


When a model trains predominantly on Global North datasets, it becomes accurate relative to those conditions. Deployed in India, the same model produces outputs that are internally coherent but contextually unusable. The system does not hallucinate. It answers the wrong question with authority.


NITI Aayog’s October 2025 report estimates that 490 million informal workers, who contribute nearly half of India’s GDP, remain outside formal systems of protection and opportunity. These are exactly the users most likely to receive AI-generated advice never designed for their conditions. A credit-scoring model trained on formal income histories will systematically misread the financial profile of a piece-rate worker in Tiruppur. An agricultural advisory system calibrated to temperate, high-rainfall farming data will produce technically correct but locally useless guidance for a smallholder farmer on the Deccan Plateau. In both cases, standard technical audits will rate the system as high-performing. In both cases, the user is worse off for having received the advice.


The problem extends beyond agriculture and credit. Health AI deployed at primary care centres in Jharkhand or Odisha encounters patients whose symptom profiles, dietary patterns, and disease burdens differ substantially from the populations on which most diagnostic models were trained. Welfare scheme delivery systems that use AI to screen eligibility routinely encounter documentation gaps and livelihood arrangements that formal datasets do not represent. In each of these domains, technical accuracy is the wrong metric. What matters is whether the output serves the decision the user actually needs to make.


Miranda Fricker’s concept of epistemic injustice describes this precisely. When a system fails to transmit knowledge because of whose knowledge it encodes, it actively diminishes the user’s capacity to make accurate decisions. This is not a secondary harm. It is a primary policy failure that India’s current audit vocabulary cannot name, let alone prevent.


Falling Short

MeitY’s November 2025 Guidelines rest on seven sutras. The fourth is Fairness and Equity. The second is People First. Both are stated as binding principles. Neither is operationalised.


The proposed institutional architecture, an AI Governance Group, a Technology and Policy Expert Committee, and an AI Safety Institute, focuses on risk classification, incident reporting, and technical testing. None of these bodies, as currently mandated, is designed to evaluate whether a model’s outputs serve the informational needs of the populations they reach. The Guidelines acknowledge India-specific risks, including linguistic exclusion and harms to vulnerable groups. They do not establish a standard which can measure those risks.


This is not a drafting oversight. It reflects a deeper assumption embedded in the architecture: that getting the technical infrastructure right will produce equitable outcomes downstream. That assumption does not hold in contexts where the data generating the infrastructure is itself unrepresentative. Fairness cannot be a sutra if the measurement standard to assess it is absent.


The RBI’s August 2025 FREE-AI Committee report moves closer. It recommends bias audits, explainability disclosures, and model risk management for financial sector AI, backed by board-level accountability and third-party audit requirements. It is also sector-specific. India deploys AI in agriculture, health, welfare scheme delivery, and credit access, where consequences of contextual failure are at least as severe, and where equivalent audit requirements do not yet exist.


AIKosh’s 7,541 datasets are a genuine policy achievement. Twelve startups have been selected to build indigenous large language models, including Sarvam AI’s 120-billion-parameter model and Soket AI’s multilingual foundation model, both designed to reflect India’s linguistic diversity. These are supply-side investments.


They do not address the demand-side question: whether the data being collected reflects the informational realities of users who are informal, rural, multilingual, and low-income. High-quality representative data for these populations is expensive to collect and curate. It is also a public good, meaning no private actor has sufficient incentive to produce it at the required scale. Market-driven AI development gravitates toward Western-centric datasets because they are cheaper and more abundant. This is a predictable consequence of data economics, not a coordination problem that goodwill or additional GPU capacity resolves.


The social costs of resulting bias accumulate in populations least able to absorb them: crop forecast errors for farmers without an insurance buffer, credit rejections for informal workers without a formal appeal mechanism, and diagnostic errors in health AI deployed at primary care centres where second opinions are unavailable. These costs are not currently measured, attributed, or reported anywhere in India’s AI governance architecture. Without a mandated mechanism to track and attribute contextual failure, the costs remain invisible to the institutions deploying the systems. An AI vendor whose model scores well on global benchmarks has no regulatory obligation to measure how it performs for a first-generation smartphone user receiving agricultural advice in Gondi or Tulu. The absence of that obligation is a policy choice with distributional consequence


Operational Changes

Amartya Sen’s Capability Approach offers the right measurement standard. Applied to AI governance, this shifts the audit question from “Is the model accurate?” to “Does this output expand what this user can know, decide, and act on?”


First, MeitY should mandate Ethical Impact Assessments for AI systems deployed in public service delivery, evaluating contextual adequacy alongside technical accuracy using benchmarks developed with the populations the systems serve. The proposed AI Safety Institute is the right body to house this function, but its mandate requires explicit expansion to include it.


Second, AI audit boards must include social scientists, development economists, and domain specialists alongside engineers. Evaluating epistemic adequacy is not a technical judgment. Interdisciplinary audit composition is the mechanism that the Guidelines currently lack.


Third, AIKosh should publish a public audit of what its 7,541 datasets represent across informal economic conditions, scheduled language speakers, and smallholder agricultural contexts, creating a procurement signal for targeted curation aligned with the NITI Aayog inclusion agenda.


Fourth, IndiaAI Foundation Model selection criteria should require contextual performance benchmarks alongside global accuracy metrics. Indigenous models that perform well on international leaderboards but poorly for scheduled language speakers in rural public health contexts have not met the mission’s own stated goals.


India’s AI governance framework has the scale and institutional architecture to do this well. What it currently measures, technical accuracy and infrastructure scale, is necessary but not sufficient. Epistemic adequacy is the standard that India’s AI for All vision requires. The analytical tools to operationalise it exist in economics, philosophy, and development policy. Adding them to the audit mandate is a governance choice, and one India is positioned to make.


(The author is an independent public policy researcher. Views personal.)

Comments

bottom of page