On AI, Narayana Murthy is both right and wrong

Infosys co-founder Narayana Murthy, often in the news  for his candid opinions ranging from 70-hour work weeks to what is perfect parenting, has one again stirred new debate. Murthy has dismissed much of India’s AI enthusiasm, and cautioned that “silly old programmes” are being marketed as AI. Murthy’s comments raise an important question — are we merely re-branding traditional software as AI, or is India truly making headway in this field?

Murthy is both right and wrong. While there is indeed some over-hyped AI discourse, India has also made remarkable strides in the field.

India was fourth in Stanford’s Global AI Vibrancy ranking, demonstrating its growing AI capabilities. The government has been actively promoting AI through initiatives such as the National AI Strategy and partnerships with major technology firms. Investments have poured in, with Microsoft announcing a $3 billion investment to expand AI and cloud capacity in India.

Besides, India is home to 16 per cent of the world’s AI talent, according to the Centre for Security and Emerging Technology (CSET). Indian universities are increasingly focusing on AI, and companies such as TCS, Infosys, and Wipro are investing heavily in up-skilling programmes to develop AI expertise. India’s AI market is projected to reach $8.3 billion by 2025 growing at a compound annual growth rate (CAGR) of 20.2 per cent.

The use of AI is expanding across industries, from healthcare to agriculture, with applications in crop prediction, fraud detection, and personalised learning platforms. Additionally, tech giants such as Nvidia, Google, and OpenAI are actively engaging with Indian researchers and businesses. Nvidia, for instance, has launched a Hindi-language AI model to enhance AI adoption in regional languages.

AI’s expanding role across industries

The use of AI is expanding across industries, from healthcare to agriculture, with innovations in crop prediction, fraud detection, and personalised learning platforms. In healthcare, AI-powered diagnostic tools such as Qure.ai’s chest X-ray analysis are helping detect diseases like tuberculosis more accurately, while AI-driven drug discovery platforms are assisting in faster identification of new treatments. In agriculture, startups like CropIn are using machine learning to help farmers optimise irrigation and predict yields, while AI-equipped drones are being deployed to monitor crop health.

Where Murthy is right

Of late India is witnessing growing instances of what has been termed as ‘AI washing’ — a marketing gimmick designed to capitalise on the hype surrounding AI. Most of the time without substantial or genuine integration of such technologies, companies are selling their products or services with an ‘AI’ tag. This is where Murthy is correct.

Many firms are now re-branding traditional software as AI, often using basic automation and rule-based programming while marketing it as cutting-edge AI. Murthy’s timely warning reflects a reality where many startups and corporations claim AI leadership without substantial AI-driven innovation. Amazon’s ‘Just Walk Out technology’ recently came into fire after it was reported that the system relied heavily on manual verification processes.

We are witnessing a rising trend where a basic chatbot programmed with pre-calibrated responses is advertised as AI-powered even though it lacks machine learning or natural language understanding capabilities. Similarly, many so-called AI chatbots in customer service are merely rule-based programmes responding to pre-set keywords, rather than true AI-driven conversational models. Even in fraud detection, some financial firms use predefined rule-based systems, which — while useful — lack the sophistication of genuine AI-driven anomaly detection.

Dearth of quality research

AI development in India cannot move beyond chatbots and recommendation engines unless there is more investment into deep learning, generative AI, and AI for scientific research. Universities and companies must prioritise high-impact AI research, collaborating with global institutions to improve research quality and practical applications.

While India has produced a vast number of AI research papers, their global impact is relatively low. A study found that India ranked 14th in AI research output, contributing just 1.4 per cent of papers in top AI conferences between 2018 and 2023. When compared to the United States and China, which lead in high-impact AI research, India still has a long way to go. 

The government admitted in Parliament that India’s research and development (R&D) budget is 0.6 per cent of the GDP — it is 3.4 per cent in the US, 2.4 per cent in China, and 5.7 per cent in Israel. If India needs to build world class models such as OpenAI’s (ChatGPT) or China's DeepSeek R1 it needs to bridge this research gap.

Overcoming AI infrastructure challenge

AI requires substantial computing power, and India still lags in developing AI-specific infrastructure such as high-performance GPUs and cloud computing capabilities. While efforts are underway to improve this, the country remains dependent on foreign technology providers for AI hardware and expertise. Unlike OpenAI, DeepMind, or Anthropic, India has yet to produce an AI startup of global influence.

While many Indian AI startups focus on niche applications, few have made breakthrough innovations that shape the global AI landscape. Developing indigenous AI chips and cloud computing capabilities will help reduce dependence on foreign technology and create a stronger AI ecosystem. AI deployment in India must also align with ethical AI principles, ensuring inclusivity, data privacy, and fairness in algorithmic decision-making.

Murthy’s criticism - a reality check

Murthy’s criticism, while blunt, offers an important reality check and it should be taken as a constructive challenge, rather than outright dismissal. On the other hand, much to Murthy’s disliking, there may be a hype surrounding AI in India, but such hype is not unique to India alone. Around the world, companies have exaggerated their AI capabilities to attract investment. Though late, India is also making slow but steady progress, and dismissing the entire sector as a collection of ‘silly old programmes’ would be an oversimplification. 

Indian industry leaders must refrain from superficial branding and focus on genuine, high-quality AI innovation. We must work towards strengthening research, infrastructure, and ethical AI development. India has the talent, resources, and market potential to become a global AI powerhouse. The challenge now is to move beyond hype, and build real AI solutions that have lasting impact. If India can navigate this transition wisely, we may soon find fewer ‘silly old programmes’ and more world-class AI breakthroughs emerging from India.

Abhishek Patni is a New Delhi-based senior journalist. X: @Abhishek_Patni.