India's CPI Inflation In May 2025 Falls To 2.82%, Lowest Since February 2019
Last Updated:
India's retail inflation in May 2025 declines to a 6-year low of 2.82%; food inflation falls to 0.99%, attributed to lower prices of pulses, vegetables, fruits, and cereals.
CPI Inflation In May 2025: India's retail inflation, based on the Consumer Price Index (CPI), in May 2025 declined to 2.82%, which is the lowest YoY inflation after February 2019, according to the latest official data released on Thursday.
The CPI inflation had stood at 3.16% in the previous month of April 2025 and 4.75% in May 2024.
'Year-on-year inflation rate based on All India Consumer Price Index (CPI) for the month of May, 2025 over May, 2024 is 2.82% (Provisional). There is decline of 34 basis points in headline inflation of May, 2025 in comparison to April, 2025. It is the lowest year-on-year inflation after February 2019," the Ministry of Statistics & Programme Implementation said in a statement on June 12.
Food inflation was at 0.99 per cent in May, significantly down from 8.69 per cent in the year-ago month, according to data of the National Statistics Office (NSO).
'The significant decline in headline inflation and food inflation during the month of May 2025 is mainly attributed to a decline in inflation of pulses & products, vegetables, fruits, cereals & products, households goods & services, sugar & confectionary and egg and the favourable base effect," NSO said in a statement.
The food inflation in May 2025 is the lowest since October 2021.
Last week, the Reserve Bank of India (RBI) revised its retail inflation projection for FY26 to 3.7%, down from its earlier estimate of 4%. This marks the lowest average retail inflation forecast by the central bank in recent years.
The downward revision came alongside a surprise 50 basis points repo rate cut, bringing the key policy rate down to 5.5%, and a 100-bps CRR reduction as the RBI signalled confidence in the easing of price pressures amid a supportive global and domestic backdrop.
However, RBI Governor Sanjay Malhotra said, 'While food inflation outlook remains soft, core inflation is expected to remain benign with easing of international commodity prices in line with the anticipated global growth slowdown. The inflation outlook for the year is being revised downwards from the earlier forecast of 4.0 per cent to 3.7 per cent."
First Published:
June 12, 2025, 16:12 IST
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Mint
an hour ago
- Mint
Bank holiday today: Are banks open or closed on Saturday, June 14? Details here
Bank holiday today: Banks across India are scheduled to remain closed today, June 14, 2025, as it is the second Saturday of the month. According to the Reserve Bank of India calendar( RBI), banks are closed on the second and fourth Saturdays and all Sundays of the month. Banks are typically closed on the weekends except on the first, third, and fifth Saturdays of a month. In addition to weekends, banks remain closed for national, regional and religious observances, depending on the state of residence. Hence, customers should note that bank holidays differ from state to state in India. Before making a visit to the bank, you should check with the local bank branch for holiday lists. The next designated bank holiday is on Sunday, June 15, 2025. In terms of festivals, banks will be closed next on Friday, June 27, 2025 in Bhubaneswar and Imphal for Ratha Yatra and Kang. Banks are scheduled to be closed for four days in the month of June 2025, apart from weekends. Here's a list of state-wise bank holidays for this month. June 7 (Saturday) — Bakrid (Id-Uz-Zuha) — All banks closed pan-India. June 11 (Wednesday) — Sant Guru Kabir Jayanti / Saga Dawa — Banks will be shut in Sikkim and Himachal Pradesh. June 27 (Friday) — Ratha Yatra / Kang (Rathajatra) — Banks closed in Odisha and Manipur. June 30 (Monday) — Remna Ni — Banks closed in Mizoram. Access to online banking services is available across the country even if the nearest bank branch is closed. Customers may utilise these services for convenient financial transactions. Requests for fund transfers can be submitted using NEFT/RTGS transfer forms, demand draft request forms, and chequebook forms. Credit cards, debit cards, and ATM cards can be accessed through card services. Additionally, services such as account maintenance forms, setting up standing instructions, and applying for a locker are also available.
Time of India
an hour ago
- Time of India
Saturday bank holiday: Are banks open or closed today on June 14, 2025?
Are banks open or closed today on June 14, 2025? Upcoming bank holidays in June 2025 Academy Empower your mind, elevate your skills Digital banking Bank holidays in June 2025 Jun-25 7 11 27 30 Agartala • Ahmedabad Aizawl • • Belapur • Bengaluru • Bhopal • Bhubaneswar • • Chandigarh • Chennai • Dehradun • Gangtok • Guwahati • Hyderabad - Andhra Pradesh • Hyderabad - Telangana • Imphal • • Itanagar Jaipur • Jammu • Kanpur • Kochi • Kohima • Kolkata • Lucknow • Mumbai • Nagpur • New Delhi • Panaji • Patna • Raipur • Ranchi • Shillong • Shimla • • Srinagar • Thiruvananthapuram • Holiday Description Day Bakri ID (Id-Uz-Zuha) 7 Sant Guru Kabir Jayanti/Saga Dawa 11 Ratha Yatra/Kang (Rathajatra) 27 Remna Ni 30 Bank holidays in India are not uniform across all states. The Reserve Bank of India (RBI) issues an official calendar each year detailing state-wise bank holidays. All banks including public, private, cooperative, and regional are closed on the second and fourth Saturdays of every month, in addition to Sundays and national or regional scheduled and non-scheduled banks will be closed today on June 14, 2025 as it is the second Saturday of the month. The first, third, and fifth Saturdays (if applicable) are regular working days unless specified in the bank holiday 27: Friday: Ratha Yatra/Kang (Rathajatra)Banks will be closed in Orissa and Manipur on June 27 for the occasion for Ratha 30: Monday: Remna NiBanks will be closed in Mizoram on June 30 for Remna N which is also known as Peace such as cash transactions, cheque deposits and clearance, account opening and loan disbursements can be done only in the branch. So such transactions should be done before or after the weekend. With internet and mobile banking, customers can carry out transactions like checking balances, transferring money, or paying bills without visiting a branch, even during weekends and holidays. Services like UPI, IMPS, and mobile wallets enable real-time fund transfers, making it easier to handle urgent payments or transfers, especially when banks are closed.
Mint
an hour ago
- Mint
India must stay nimble as ‘old multilateral architecture' breaks down, says Sanjeev Sanyal
New Delhi: Despite the ongoing global turbulence, India's economic growth remains robust, said Sanjeev Sanyal, a member of the PM's Economic Advisory Council. However, as the US and China shape a world increasingly split between two distinct supply chains, India must adapt swiftly and act boldly, with American reshoring offering a potential advantage, he told Mint. He added that with limited immigration, openings in services such as IT and consulting will grow, and India is well-placed to fill the gap. At home, though, falling birth rates pose a major risk as India is 15 years past its peak in terms of annual births, leading to shrinking classrooms. Indians are simply not having enough babies, Sanyal said, adding that the shift is nationwide and demands urgent adjustment. He also emphasised the need to reform and improve systems such as Employees' Provident Fund Organisation (EPFO) and Investor Education and Protection Fund Authority (IEPFA). Here are some edited excerpts from the interview. We meet at a time of significant global churn, with unfolding events likely to impact the economy. Meanwhile, India continues to post robust growth, currently estimated at 6.5%. Where is this trajectory headed? From your vantage point, how do you see the Indian economy shaping up? And given the external factors at play, how sustainable and robust is this growth likely to be? Sanyal: I think the Indian economy, despite all the turbulence we have seen, remains very robust in terms of macroeconomic stability. Inflation, as you can see, is now around 3%, and wholesale inflation is even lower. (Consumer Price Index-based inflation eased to 2.82% year-on-year in May, down from 3.16%, according to official data released on 12 June, after the interview). Across almost every macroeconomic parameter, we are well within control. The financial system is strong. Banks are well-capitalised and flush with liquidity, especially after the cash reserve ratio (CRR) cut. The external accounts are in decent shape, and India is not a heavily indebted country. Our debt-to-GDP ratio is also in a reasonable range. So, by virtually any macroeconomic stability measure, we are in good shape. Growth, too, is solid. A 6.5% GDP growth rate is quite strong, particularly when much of the world is not performing well. So, I would say we are doing well in terms of delivering stable, solid growth. That said, we should aspire to grow even faster. The question is: should we aim to push growth beyond 7%? I believe we exercised appropriate caution over the past year, and that was the right call. We have been steadily reducing the fiscal deficit, from well over 5% to 4.8%, and we intend to bring it down further this year. Also read: India concerned about supply disruptions in Strait of Hormuz as oil prices surge Given this trajectory, and with inflation now under control, I think the Reserve Bank did the right thing by easing policy quite aggressively. It wasn't just the 50 basis point rate cut, there was also a 100 basis point CRR reduction. This shows we are now ready to start pressing the accelerator. So far, we have kept our foot only lightly on it; now we're beginning to push a little more. That doesn't mean we should slam it down. The road ahead still has plenty of potholes, so we need to remain cautious. But I believe we are now in a position to push growth higher. And we must not give up the ambition of achieving more than 7% growth. Wherever the environment allows, we should aim to exceed that threshold. In the ongoing trade talks with the US, agriculture is often cited as a key area where they seek greater access. In your view, what should be the way forward? Sanyal: It's very clear that the old multilateral architecture, such as the WTO, has essentially broken down. While recent events, like President Trump's imposition of tariffs, drew widespread attention, the system has been deteriorating for quite some time. In many ways, it was inevitable that we would reach this point. The question now is: what happens going forward? It appears we are moving towards the creation of two parallel supply chains. These will not be neat or entirely separate, they will interact with each other, but one will be largely oriented around China, and the other around the United States. That is the direction in which we are headed. As I mentioned, these systems will not be cleanly divided. They will continue to trade with each other. However, India is integrating more closely with the Western system. We already have free trade agreements with Japan, Australia, and the UAE. We have signed one with the UK and are in advanced discussions with both the European Union and the United States. Also read: Israel's war on Iran to hit Indian workforce Of course, there are specific issues to negotiate in each of these talks, agriculture being a prominent one, but the broader point is that by this time next year, it's very likely that we will have some form of trade agreement with both the EU and the US. Even if not comprehensive, these agreements are expected to be significant. This would place India at the heart of global supply chains in a way it has never been before. Another important point is that India's national attributes are quite different from those of its major trading partners, which is a good thing. Trade between similar economies offers gains through specialisation, but trade between dissimilar economies can yield even larger advantages. This is an opportunity for us to expand initiatives like 'Make in India' and grow our manufacturing footprint in targeted sectors. Of course, we are already a major player in services, but there is significant headroom to scale that up further. Also, consider that some policy choices being made in advanced economies can work to our benefit. For example, the United States is pushing to revive domestic manufacturing. Some see this as a threat to Indian manufacturing, but that's not necessarily the case. Let's say the US does manage to bring back manufacturing in certain areas. Given that it is not allowing large numbers of immigrants, legally or otherwise, it will have to redirect its existing workforce toward these new roles. This creates gaps in other sectors, such as accounting, consulting, software development and other services. Who will fill those gaps? Those tasks will need to be outsourced, and India is well-positioned to capture that work. Even if, hypothetically, the US succeeds in manufacturing something like Apple phones domestically, despite high wage costs or even with the help of automation, there will still be a reallocation of labour. The workers building those phones will no longer be doing their previous jobs. Those vacated roles will still need to be filled, and not all of them can be handled by AI, contrary to popular belief. AI is a tool, just like a hammer. And like any tool, it still needs someone to use it effectively. The net result is this: structural shifts in global labour and supply chains will open up new spaces that India can occupy. What is important is that we keep our eyes open to these emerging opportunities. Indian policymakers, businesses, and professionals need to remain flexible and adaptive because the opportunities that lie ahead are incredibly promising. With the US and China nearing a trade deal, how might such an agreement impact the (West's) China Plus One strategy? Sanyal: There is a narrow way to look at it, focusing purely on whether or not the US and China strike a trade deal, and how that affects the China Plus One strategy. I can't predict the specifics of any potential deal; we will have to adapt to whatever shape it takes. But I believe that misses the larger point entirely. What is unfolding is a fundamental shift in global power dynamics, something both China and the US understand very well. Regardless of the details, any trade deal between them is likely to be inherently unstable. In my view, India should focus on building its strengths, independent of the nature or outcome of any US-China trade agreement. This is a long-term struggle between great powers and not just about trade. At its core, this is a civilizational and geostrategic contest, with trade being only one part of a much larger picture. We should focus on our strengths rather than try to predict how it will turn out. Given the Budget announcements, particularly the revised tax slabs and concessions, is there now a case for a GST cut? Sanyal: Well, more than a cut, I think it is a simplification that has been introduced. And since we are sitting in the office of the late Bibek Debroy, let me take a moment to remember him. He consistently advocated for a reduction in the number of tax slabs, and I agree with him. In particular, the middle two GST slabs—12% and 18%—should ideally be merged into a single slab, which would significantly simplify the system. I have long believed that the system needs to be streamlined. Additionally, I think we should return to a more liberal approach regarding processes like registration. Yes, there have been instances of misuse in the past, but I believe we now have a much better understanding and oversight of the system. This puts us in a position to adopt a more liberal, trust-based framework for registration and similar compliance requirements. Given the government's economic outlook and its understanding that states must be taken on board, with decisions to be made by the GST Council, what is the thinking on this matter? Sanyal: Generally speaking, the central government has actually been quite liberal on these matters. However, it's important to remember that the states must be taken along in the process. Contrary to popular belief, GST is not solely run by the central government. We have been advocating for simplification for some time now. As I mentioned, the late Bibek Debroy consistently believed that we need to radically simplify the system. One of the government's recurring concerns is that private-sector capex hasn't picked up despite policy support and repeated discussions. Is it time for the industry to step up and take more initiative on its own? Sanyal: Yes, the private sector needs to step up risk-taking. This is where we need to understand that creative destruction is something to be embraced. Let me illustrate this. Take a look at the top 10 listed companies by market capitalisation in the US, China, and India. Now compare today's list with that from 20 years ago. In the US, except for Microsoft, nine of the top 10 have changed. Companies like Meta, Alphabet, and Tesla were either tiny or barely known two decades ago. In China, while three of the top 10 are still state-owned banks, the remaining seven have changed. Some of these, like BYD, were virtually unknown even just a few years ago. Yet, today, they rank among China's biggest companies. Now look at India. Our top 10 or even top 20 companies have barely changed. Names like Reliance, Infosys, TCS and State Bank of India still dominate the list. There are very few new entrants. In other words, there hasn't been enough churn. There hasn't been enough creative destruction. This is because, as a society, we don't sufficiently value the churn of ideas, companies and the spirit of risk-taking. Also read | Trump-induced slowdown: Can it be reversed? And this brings me back to a broader point: where are our wild, bold ideas? In our ancient past, we celebrated unconventional thinkers and encouraged them. Today, we tend to view such people as unsettling or disruptive. Instead of fostering their energy, we often suppress it. But real progress depends on embracing those ideas, and the disruption that often comes with them. In FY25 India grew at 6.5%, down from 9.2% in the previous fiscal year. Is the current growth rate sufficient to make India a developed nation by 2047? Sanyal: Essentially, we want to grow at the fastest rate possible without jeopardising our macroeconomic stability. Rapid growth is desirable, of course, but it's important to remember that we must avoid the pitfalls experienced by the East Asian Tigers. They grew very rapidly for a period, but their financial systems collapsed during the Asian financial crisis, and several of them never regained that growth momentum. We don't want to end up in a similar situation. Sustaining growth is just as important as achieving high growth. It's worth noting that after the global financial crisis of 2007, we did attempt to accelerate growth. While it gave us a short-term boost, it also led to a breakdown in our financial system, placing India among the so-called 'Fragile Five' by 2013. We then spent several years cleaning up that mess. We must avoid repeating that mistake. So, the first point is: yes, we should aim to grow as fast as possible, but never at the cost of macroeconomic or financial stability. Now, having made that caveat, should we grow faster than 6.5%? Yes, absolutely. And do we have the room to push a little harder right now? Yes, we do. Inflation has come down, and there are no significant pressures on the system at present. In response, we have already seen monetary authorities cut interest rates sharply just last week. That move injects liquidity into the system and provides monetary stimulus. We now have to see how the economy responds to it. In the coming months, we will assess whether that measure is sufficient, or if additional steps are needed. Ultimately, that will depend on evolving circumstances. The four-day conflict with Pakistan has naturally brought national security back into sharp focus. In this context, and given the demands on our defence preparedness and spending, is there a compelling case to raise defence expenditure to at least 2.5% or even 3% of GDP? Sanyal: The point is, just as I said we shouldn't give up our ambition for high growth simply because we occasionally find ourselves in turbulent times, by the same logic, we should aspire to have world-class defence systems. But that doesn't mean we go out and begin spending blindly. The question is: what can be spent, and what is useful to spend on? I don't believe the lack of resources is currently a binding constraint. We have repeatedly awarded large contracts, both to Indian companies and, at times, to foreign firms. Over the past decade, there has been a significant expansion of our domestic defence capabilities, and I don't think anyone doubts that. We are even exporting now. So, resource availability is not the issue. The real question is how fast the Indian industry, both public and private, can absorb these resources and scale up. There are constraints. For instance, we can build the entire Tejas aircraft, but we still depend on external suppliers for the engine. We are now trying to develop our own, but until that's fully operational, we remain dependent. Also read | Trump's crackdown on migrants: Will the American dream turn sour for Indians? The larger point is that we must respond to the situation as it exists, not as we wish it to be. That's why I often say I have an allergy to rigid, hard plans. You can't operate like that. The goal is to have clear objectives, but the achievement of those objectives requires the flexibility to adapt to current realities. For example, if our objective is to become a developed country by 2047, the path to getting there depends on how well we respond, flexibly and effectively, to the opportunities and challenges of the moment. The IMF expects India to surpass Japan as the world's fourth-largest economy in 2025. But why do we continue to lag significantly on key indicators like per-capita income? Sanyal: I have never really understood this discussion – why are we even debating it? Of course, we are a poor country. We may be the fourth-largest economy by GDP and have the largest population, but that doesn't change the fact that we're still poor on a per-capita basis. So every time it's mentioned that we have become the world's fourth- or third-largest economy, someone inevitably says, 'But we are still a poor country." Yes, we are, and we want to become less poor. That's precisely why we need to grow our GDP. These two facts are not in conflict. Per-capita income is simply GDP divided by population. You cannot increase per-capita GDP without increasing overall GDP. How is the issue of declining fertility rates, especially in southern India, becoming a political flashpoint, and what are its implications for the region's demographics? Sanyal: This is an issue the country as a whole must address. Our national fertility rate is declining, not just in the southern states, but across the entire country. If it weren't for Bihar and Uttar Pradesh, we would be far below the replacement rate. Even now, we are already below replacement levels, and without those states, the situation would be far worse. Interestingly, states like West Bengal have fertility rates even lower than those in the south. This is a serious national concern that requires urgent attention. Oddly enough, we still have population control departments operating across the country. These are outdated and must be shut down, even in Bihar and Uttar Pradesh. Not only are birth rates declining in those states, too, but we are increasingly reliant on them to keep our national fertility levels at a reasonable threshold. Rather than criticising Bihar and Uttar Pradesh for their relatively higher birth rates, other states should now be thinking about how to increase theirs. If we don't act now, we risk facing the same demographic crisis as China and other East Asian economies. I strongly advocate for the immediate closure of all population control departments across India. Are these conversations happening in the government? Sanyal: Some awareness is building, but I believe we need a much broader public debate on the issue. While some policymakers are more cognisant of it, I still routinely hear from the general public: 'Oh, India has a population problem. We need to do something about it." That is no longer true. India's population is growing now primarily because we are living longer, not because we are having more children. We are already 15 years past our peak in terms of annual births. The real issue is that we are not having enough babies. In several parts of the country, schools are already being shut down due to falling student numbers. And this trend isn't limited to southern India, it's also happening in states like Himachal Pradesh and others. While some closures are taking place, we need to carry them out more routinely and systematically. Whenever I mention school closures, it tends to trigger emotional responses. But the reality is, in many regions, now more than half the country, schools simply don't have enough children to remain viable. Classrooms are shrinking, and resources are being spread thin. We must start adjusting to the fact that birth rates in India are declining rapidly. Which are the most problematic states in this regard? Sanyal: Well, some southern states have very low birth rates. But West Bengal also has a remarkably low birth rate. Urban Bengal, particularly Kolkata, has birth rates that resemble those of South Korea. If I recall correctly, Kolkata's fertility rate is around 1.2. I have written about this in the past. That's extremely low, alarmingly so, and it will create serious challenges for us if we don't address it and begin thinking seriously about population dynamics. We keep repeating, 'Oh, we are a young country." No, we won't be for long. In 20 years, by the time we hope to achieve Viksit Bharat in 2047, we will already be an ageing country. Whether or not we reach developed-country status by then is up for debate. But what is not up for debate is the fact that our workforce for 2047 is already born, and the demographic transition is underway. Should Indian states with low birth rates actively pursue this, especially after the problems faced by China? Sanyal: This is precisely why demographics is such a powerful force, and why it will pose a serious long-term problem for China. Even if China were to attempt a reversal today, say, by somehow convincing every Chinese woman to have three children starting tomorrow, the impact on their workforce would only be felt after 25 years or more. A child born today will not enter the workforce until their early 20s, around 2050. Meanwhile, to care for this hypothetical baby boom, China would need to withdraw a significant portion of its existing female workforce, along with doctors, nurses, and other professionals, from productive sectors like manufacturing or services to focus on child-rearing and related needs. So, paradoxically, in trying to address a shrinking workforce, they would put even more pressure on it in the short term. The irony is stark: any effort China makes to fix its demographic crisis will likely worsen the situation before it gets better. Also read: Rush to beat tariffs boosted India's exports to US, but what next? But the point I am making is this: we in India are also heading in the same direction. Instead of complaining about higher birth rates in Bihar and Uttar Pradesh, other states should seriously consider how to raise theirs. And let me add, this is an extraordinarily difficult thing to achieve. No country in the world has yet been able to reverse its declining fertility rate in a sustained and meaningful way. This is a matter that demands national attention and a broad public debate. Which upcoming reforms do you believe could have the most significant impact on the economy? Sanyal: The first thing to understand is that we do need to undertake certain structural reforms. However, most of the reforms we now need are process reforms. These aren't large, sweeping changes like GST that affect everyone; rather, they consist of numerous small to medium-sized reforms. These reforms may seem minor in isolation, and the general public may not take much notice, but they are crucial for the specific sectors they impact. For instance, we need to improve the functioning of systems like the Employees' Provident Fund Organisation (EPFO) and the Investor Education and Protection Fund Authority (IEPFA). This has been a longstanding issue, and while I have written about it and am personally working on IEPFA processes, it's clear these systems need to be streamlined. There are also many issues related to ease of doing business and ease of living that require attention. Take our outdated building codes, many of which are stuck in the 1950s. These not only impede development but also result in the kind of architectural monstrosities we see around us today. Addressing such challenges requires meticulous process reforms, fixing one system, one regulation at a time. Additionally, whether at the central or state level, numerous defunct government bodies need to be shut down. But this must be done systematically. What about DOGE? Sanyal: The example of Elon Musk's approach with DOGE (Department of Government Efficiency) is a cautionary tale. While I have great respect for him in other domains, his method of cutting back was haphazard. Shutting down something like USAID (United States Agency For International Development) might have been justified, but without a sustained, strategic approach, such efforts can backfire. There's a real risk that the failure of such experiments, like DOGE, could undermine the broader case for improving government efficiency globally. Is there a case for a department like DOGE in India? Sanyal: Perhaps. In any case, we are already undertaking several initiatives here in India. For example, I am personally involved, as part of the PM's Economic Advisory Council, in driving process reforms in a significant way. You may have read about some of these efforts, including in your publication. Whether it is through the proposed Deregulation Commission or other mechanisms, this kind of reform requires sustained effort. A person or group must remain embedded in the system for a long period, working steadily, one reform at a time. There is no shortcut to this process. In your paper last year, you called AI one of the biggest challenges and opportunities of our time and advocated for a dedicated regulator. How do you see the AI landscape evolving, and what kind of regulatory framework do you envision? Sanyal: We will increasingly be handing over significant aspects of our lives to AI-based systems – this is already happening. As a result, there is a real risk, because we are entrusting autonomous decision-making to systems that may have unintended consequences, especially when interconnected. Beyond concerns of malicious use, such as manipulation or sabotage, even well-intentioned AI systems can behave unpredictably. This makes the case for some form of regulation compelling. Broadly, there are two dominant schools of thought on AI regulation. One is what we might call the US approach, and the other is the continental European model. I am simplifying, but the US approach largely favours self-regulation by companies, with the government stepping in reactively when problems arise. This approach has its merits, especially given the pace of innovation. Self-regulation often does not work. Also read | US in stagflationary shock, India remains outlier: BofA The European approach, by contrast, assumes that bureaucrats can proactively identify and regulate risks — that they can anticipate the dangers posed by specific AI applications or technologies in advance. But this raises a fundamental question: if these regulators in Brussels are so good at anticipating technological risks, why are they not the ones developing the next generation of AI? The core issue is this: effective regulation cannot depend on a regulator's ability to predict the future of an evolving technology. What will the Indian model be like? Sanyal: I will come to that. The point is, that the American model has the advantage of recognising that regulators and bureaucrats cannot predict the future. However, laissez-faire has its challenges. The US relies on something called tort law – essentially, if a company causes harm it faces massive penalties, often running into billions of dollars, potentially destroying the company or the individuals responsible. I am not suggesting this is necessarily the best system, but it does put skin in the game for those taking risks. That said, both the American and European approaches have serious limitations. Self-regulation, as in the US, often fails because large corporations have powerful legal teams that help them evade accountability under Tort law. Moreover, tort law is reactive — it steps in after the damage is done. The European model, on the other hand, is also almost certain to fail. It relies on the ability of bureaucrats to foresee the risks of AI systems in advance. But this either leads to overregulation — stifling innovation entirely — or underregulation, where risks are underestimated and things spiral out of control. When something goes wrong, companies can just say, 'But you gave us permission." So both approaches are flawed. In India's case, I have proposed a completely different model, one that comes with its trade-offs. My suggestion is: do not attempt AI regulation based on the ability to predict evolution. Instead, deliberately compartmentalise AI systems from the outset. There should be strict silos. For example, the AI that runs banking systems must be completely disconnected from the AI running satellites or the electricity grid. Also read | Mary Meeker's AI report: Decoding what it signals for India's tech future To give you a real-world example, in June last year, a single coding error in a Microsoft system caused disruptions across airports, power grids, and other critical infrastructure. And that was with static code. Now imagine this kind of chaos with dynamic, self-learning AI systems. So, my first recommendation is to create hard boundaries between AI systems, even if that causes some inefficiency. Everything does not need to be connected to everything else. Second, there must be legal requirements for human overrides at every critical control point. Third — and most important — we must institute explainability audits. We already manage similar complex adaptive systems in this way, like financial markets. Sebi, for instance, doesn't need to predict where the stock market is headed to regulate it. Similarly, an AI regulator doesn't need to know where AI is going, because if someone truly knew that, they'd be building companies, not regulating them. So, the regulator must be agnostic. Just as we don't appoint the best investor to head Sebi, we don't need the top AI scientist to run AI regulation. What we do need are mechanisms such as explainability audits. AI systems must be able to justify their decisions. [We need] mandatory human overrides, legal requirements for human-in-the-loop oversight, clearly defined shutdown protocols, and compartmentalisation. AI systems should be designed so they cannot cascade into each other and cause systemic failure. These already exist in financial markets: circuit breakers, transparency requirements, Chinese firewalls, and related-party transaction rules are all tools designed to manage systemic risk. Why do we need this for AI? Because, like financial markets, AI is a complex adaptive system. And therefore, it should be regulated with the same logic. It's really that simple.
