How societies control and regulate AI will determine how powerful it becomes

Air Cmde T.K. Chatterjee (retd)

Artificial Intelligence (AI) is the buzzword of our time and has become one of the most important forces shaping the modern world. Once seen mainly as a technological innovation, it is now influencing economics, security, communication, and governance on a global scale. Governments, companies, and international organisations increasingly view AI not only as a tool for efficiency and innovation but also as a source of strategic power. 

As a result, AI is no longer confined to laboratories or commercial platforms; it has become deeply connected to questions of national interest, global competition, and international order. In this context, understanding AI is essential to understanding how power is changing in the 21st century. India is an emerging player in this field, a swing state between the US and China in technology, and a key player in the digital future of the global South.

Effect on Geopolitics

AI is transforming geopolitics by changing how states compete for power, security, and influence. One of the clearest examples is the growing rivalry between the United States (US) and China over the computing infrastructure that underpins AI, especially advanced semiconductors. The US has imposed export controls on high-end AI chips and semiconductor manufacturing equipment to restrict China’s access to the computing power needed for frontier AI systems, while China has responded by accelerating its drive for technological self-reliance in chips, data centres, and AI infrastructure.

China has also used its dominant position in critical minerals and rare earth processing to counter US pressure: export controls on materials such as gallium, germanium, graphite, and several heavy rare earth elements have shown how control over supply chains can be used as leverage in response to US tariffs and semiconductor restrictions. This rivalry reflects the wider belief that control over computing capacity and strategic inputs is central to geopolitical power. As the White House states in America’s AI Action Plan (23 July 2025), AI breakthroughs could ‘reshape the global balance of power,’ while Chinese Premier Xi Jinping stated in a Xinhua report published on 29 April 2025 that China must ‘achieve self-reliance and strength’ in AI. As I write, CNN reports, ‘China has clinched the top spot on a list of the world’s most powerful supercomputers, overtaking the United States for the first time since 2017 with a model powered by homegrown chips amid an intense race for tech supremacy between the two superpowers.’ The point to note is the catch phrase—‘homegrown chips’.

India’s position in this contest is distinctive: it has significant strength in chip design, engineering talent, and back-end assembly, testing, marking, and packaging, but it still depends heavily on foreign fabrication capacity for advanced chips. This means that India has an important place in the semiconductor supply chain, yet its strategic autonomy in AI remains constrained until its domestic fabrication ecosystem matures. Recognising this gap, the Indian government has launched a strong policy push through the India Semiconductor Mission and related incentive schemes to build semiconductor fabrication, packaging, and design capacity at home; as the Press Information Bureau (PIB) noted in its backgrounder India Semiconductor Mission 2.0: A Major Push in Budget 2026 towards Semiconductor Self-Reliance (7 February 2026), this effort is intended to deepen domestic capabilities across the semiconductor value chain. In addition, AI is influencing diplomacy and governance through the concept of digital sovereignty, whereby states seek to retain control over sensitive data and critical digital systems rather than relying excessively on foreign companies; India, for example, has advanced this goal through the Digital Personal Data Protection Act 2023, payment data localisation rules, and the development of domestic digital infrastructure such as UPI and DigiLocker.

On 26 August 2025, Antonio Gueterres welcoming the UN General Assembly’s decision to create an Independent International Scientific Panel on AI and a Global Dialogue on AI Governance, he said these mechanisms would help ‘build on the Global Digital Compact’ by providing a ‘significant step forward in global efforts to harness the benefits of artificial intelligence while addressing its risks,’ so that AI governance is grounded in shared scientific evidence rather than competing narratives. On 23 September 2025, briefing the Security Council on AI, he warned that ‘artificial intelligence holds vast potential but poses grave risks if left unregulated,’ stressing that ‘AI is no longer a distant horizon—it is here, transforming daily life, the information space, and the global economy at breathtaking speed,’ and insisting that ‘humanity’s fate cannot be left to an algorithm,’ as he reiterated his call to ban lethal autonomous weapons operating without human control.

Effect on Military Strategies

AI is having a particularly strong effect on military strategy. It increases the speed at which armies can collect information, assess threats, and respond on the battlefield. In earlier forms of warfare, commanders often relied on slower chains of reconnaissance and human analysis, which delayed decision-making. 

By contrast, AI can process large volumes of satellite images, drone footage, signal data, and battlefield reports in real time, helping militaries detect patterns and identify targets much faster. A 2024 study in the Journal of Global Security Studies, published by Oxford Academic on 8 May 2024, argues that the most important military use of AI so far is not fully autonomous weapons, but the way it improves data processing, intelligence, and targeting. This is strategically significant because it shortens the gap between detection and action, giving forces a better chance to strike first, defend more effectively, or adapt quickly to changing battlefield conditions. In this sense, AI is reshaping military strategy by making warfare more data-driven, faster, and increasingly dependent on information superiority rather than on numbers and firepower alone.

AI is also changing military strategy by transforming the structure and conduct of war itself. The conflict in Ukraine has shown how AI-assisted drones, surveillance systems, and software-based targeting tools can help a weaker force offset an opponent’s advantages in manpower and equipment. Reports on the war describe a shift toward semi-autonomous drones that can continue their final approach even in jammed environments, reducing reliance on constant human control and making strikes more precise under electronic warfare conditions. 

For example, IEEE Spectrum reported on 9 April 2025 that Ukrainian drone developers were using AI-based navigation systems to keep drones on course even when radio and satellite links were jammed. This changes military planning because armies must now think not only about tanks, aircraft, and artillery, but also about software, communications resilience, anti-jamming systems, and mass drone production. AI is further influencing logistics and command structures by improving predictive maintenance, supply coordination, and battlefield awareness, all of which are essential in long, high-intensity wars. Nato’s revised AI strategy, published on 10 July 2024, also reflects this shift, stating that it is ‘vital’ to use these technologies where applicable and linking AI to interoperability, capability development, and defence readiness. Overall, AI is pushing military strategy away from purely traditional models of force and toward a form of warfare centred on networks, autonomy, adaptation, and rapid innovation.

A further military issue is the rise of autonomous weapon systems, which are strategically important because they could allow faster target selection and strike decisions with less direct human involvement. Supporters argue that such systems may increase precision and reduce risks to soldiers, especially in high-speed environments where human reaction time is limited. However, they also raise serious concerns about accountability, civilian protection, and escalation. In their joint appeal published by the UN and the International Committee of the Red Cross on 5 October 2023, Guterres and Mirjana Spoljaric warned that ‘human control must be retained in life and death decisions’ and that machines with the power to kill without human involvement should be prohibited by international law. This debate matters for military strategy because autonomous weapons could lower the threshold for the use of force, accelerate battlefield decisions beyond meaningful human judgement, and make it more difficult to assign responsibility when unlawful harm occurs. For this reason, autonomous weapon systems are not only a technological development but also a major strategic and legal challenge for the future of warfare.

A recent controversy involving the Pentagon and the AI company Anthropic further illustrates the ethical and strategic tensions surrounding military uses of AI. In February 2026, Anthropic CEO Dario Amodei refused Pentagon demands to remove safeguards that prevented the company’s models from being used for mass domestic surveillance and fully autonomous weapons. As CNBC reported on 26 February 2026, Amodei said that Anthropic could not ‘in good conscience accede’ to a request for unrestricted military use of its systems. In an interview with CBS News published on 1 March 2026, he argued that current AI systems do not show the judgement of a human soldier and are therefore not reliable enough for lethal decisions without human involvement. This episode is important because it shows that the debate over military AI is not only about improving battlefield effectiveness, but also about who sets the ethical limits of state power. It raises deeper questions about democratic accountability, human control over lethal force, and whether private technology firms should be able to refuse certain forms of military application on moral grounds.

Economic Effects

AI is also having a major effect on economics by increasing productivity, changing labour markets, and redirecting global investment toward digital infrastructure. Recent research summarized by the International Center for Law & Economics in AI, Productivity, and Labor Markets: A Review of the Empirical Evidence (5 February 2026) shows that AI can significantly reduce the time needed for writing, coding, customer support, and other knowledge-based tasks, while often improving quality at the same time. This means that AI is not only creating new products and services but also changing how efficiently existing industries operate. A useful business example is the high-end travel and hospitality company, Club Med, which has used AI as part of its digital transformation by building a large data platform, introducing conversational booking tools such as GM Copilot on WhatsApp, and automating part of its customer interactions across multiple markets. That way, they could move away from a network of travel agencies spread across the world. 

This shows how AI can improve service efficiency, personalise customer experience, and open new commercial channels in service industries. While Club Med did a customer-facing service transformation, a different example is Walmart, which has used AI to strengthen back-end operational and supply chain transformation. It improved its demand forecasting, inventory control, and logistics optimisation across its supply chain, showing how digital transformation can reduce inefficiencies and improve resilience in large retail networks. 

India has also begun to attract substantial foreign investment linked to AI development. For example, Microsoft has announced multi-billion-dollar investments in India’s AI and cloud ecosystem and partnered with IndiaAI to support foundational models, datasets, and innovation centres, while Google has announced a major AI hub in Visakhapatnam. These developments suggest that India is becoming an important destination not only for digital infrastructure investment but also for AI capability-building and innovation.

Another major economic issue is the growing energy requirement of AI data centres. Training and running advanced AI systems depend on vast computing infrastructure, and this sharply increases electricity demand because data centres must power thousands of high-performance chips as well as cooling and network systems around the clock. The International Energy Agency stated in its report Energy and AI, published on 10 April 2025, that global electricity demand from data centres is projected to more than double by 2030 to around 945 terawatt-hours, with AI being the most significant driver of this increase. This matters economically because access to reliable and affordable electricity is now becoming a strategic condition for AI growth. It also means that countries seeking to lead in AI must invest not only in chips and cloud platforms, but also in power generation, transmission networks, and energy security. In this sense, the AI boom is creating a new link between digital competitiveness and the physical electricity infrastructure.

At the same time, while AI can bring enormous economic benefits, these benefits are presently unevenly distributed. According to the IMF Working Paper The Global Impact of AI: Mind the Gap (10 April 2025), advanced economies are likely to gain far more from AI than low-income countries because they have stronger digital infrastructure, more skilled labour, and greater access to data and computing power. AI is also reshaping patterns of global investment. The Stanford HAI 2025 AI Index Report, released on 7 April 2025, reports that private AI investment in the US reached USD109.1 billion in 2024, far ahead of China and the United Kingdom (UK). This shows how AI is concentrating capital in countries and firms that control chips, cloud systems, and data centres. In addition, the Federal Reserve Board noted in The Global Trade Effects of the AI Infrastructure Boom (13 February 2026) that the AI infrastructure boom is already affecting international trade by increasing demand for critical components used in data centres. Therefore, the economic effect of AI is not limited to automation alone; it is restructuring productivity, employment, investment, trade, and the global hierarchy of technological advantage.

Social Effects

AI also has the potential to improve the lives of weaker and more disadvantaged sections of society if it is made accessible, affordable, and inclusive. In education, AI tutors and translation tools can help students from poor or rural backgrounds learn in their own language and at their own pace. In healthcare, AI can support early diagnosis, telemedicine, and basic medical advice in areas where doctors are scarce. It can also help small farmers, informal workers, and micro-businesses by providing information on weather, market prices, crop diseases, bookkeeping, and customer communication. For persons with disabilities, AI-based speech-to-text, text-to-speech, and visual assistance tools can increase independence and participation in daily life. 

At the same time, AI can make it easier for citizens to access welfare schemes, public services, and financial tools through multilingual digital assistants. A concrete example comes from India, where AI-supported maternal health programmes developed with the non-profit ARMMAN and Google DeepMind have been used to identify women at risk of dropping out of care and to personalise outreach calls, helping low-income mothers in underserved communities stay connected to preventive health services. As reported by The Hindu BusinessLine on 2 July 2025, this use of predictive models shows how AI can support public health outreach for vulnerable groups rather than only serving commercial or military purposes. However, these benefits will only be meaningful if AI is governed fairly, so that vulnerable groups are not excluded by cost, bias, poor digital access, or intrusive surveillance.

Regulatory Oversight

Regulatory oversight of AI is developing rapidly, but it remains uneven and fragmented across major stakeholders. The European Union has taken the strongest formal action through the EU AI Act, Regulation (EU) 2024/1689,  which establishes a risk-based legal framework, bans certain unacceptable uses, and imposes obligations on high-risk and general-purpose AI systems. The US has acted through executive orders, agency guidance, defence-related controls, and state-level legislation, but it has not yet adopted a single comprehensive federal law, leaving oversight dispersed across different institutions. 

China has moved assertively through state-led regulation of algorithms, deep synthesis, generative AI services, and AI-generated content, yet its system is more focused on control, security, and political stability than on open global governance. India has taken some important steps in digital governance, data protection, and public digital infrastructure, but it still lacks a fully developed and sector-wide AI regulatory framework. International organisations such as the UN, Organisation for Economic Co-operation and Development (OECD), G7, and the Council of Europe have helped to shape norms, ethical principles, and treaty-based discussions, but they still lack strong global enforcement mechanisms. Private technology firms have also become major stakeholders by adopting internal safety policies, model restrictions, and ethical red lines, yet many companies still resist external regulation, remain opaque about training data and risk testing, or move faster than public institutions can respond. 

In particular, US-based hyperscalers such as Microsoft, Google, Amazon, Meta, Anthropic, and OpenAI increasingly exercise a form of quasi-sovereign power because their control over frontier models, cloud infrastructure, and compute capacity gives them influence that can rival that of states, even as governments attempt to reassert regulatory authority over AI development and deployment. Overall, the present picture is one of significant action, but also of major inaction: regulation exists, but it is incomplete, inconsistent, and still struggling to keep pace with the speed of AI development.

Conclusion

AI is not merely a technological innovation but a transformative force that is reshaping the world in multiple interconnected ways. As this essay attempts to show, AI is altering geopolitics by changing the balance of power between states, transforming military strategy through data, autonomy, and faster decision-making, and restructuring economies by increasing productivity, redirecting investment, and creating new demands on infrastructure such as energy and data centres. It also has important social implications, since it can widen access to education, healthcare, public services, and economic opportunities for disadvantaged groups if it is governed inclusively. At the same time, the growing importance of regulatory oversight shows that the future of AI will depend not only on innovation but also on the ability of states, institutions, and companies to create effective safeguards, accountability, and fair governance. Yet these benefits are accompanied by serious risks involving inequality, surveillance, accountability, and conflict. The overall significance of AI, therefore, lies not only in what it can do, but in how societies choose to regulate, distribute, and control it in the years ahead.

(Writer’s Note: Yes, I used AI to conduct the research for this article. It came up with a huge amount of information sources for each of my queries, and there were hundreds of them. I had to sift through them to find what I was looking for. I would have required a battery of assistants to do this task manually, using just Google Search, and it would have taken me months. For laypeople, technology has evolved from simple Google Search to tools like Grok, Perplexity, Gemini, Copilot, and ChatGPT. Gemini on my smartphone thinks for me while I walk my granddaughter to her school and lists the groceries I need to buy on my way home. But the decision to buy or not is my own. That is the essence of using AI. It will throw up data and options, but what to use, how to use it, how much to use, or whether to use it at all will be a human decision. As long as that remains, ethical limits and accountability will be the guardrails. If that goes away… then God help us)