Subscribe
AI is poised to become a core capability in India’s power generation, with adoption expected to rise 65 per cent by 2030, up from 35 per cent today, an ENCIS survey shows.
Covering 250 sector professionals, it finds efficiency and clean energy goals driving investment, with predictive maintenance, grid management and renewable integration seen as key applications amid rapid data centre growth.
Artificial intelligence (AI) is set to become a core operating capability in India’s power generation sector, with adoption in power plants and generation assets projected to rise 65 per cent by 2030, according to a new survey.
Around 35 per cent of Indian power generation firms have already adopted AI in their core operations, the survey by think tank Energy & Climate Initiatives Society (ENCIS) found. The findings come as India’s electricity system expands to meet rising power demand while adding large amounts of renewable energy and digital infrastructure, increasing the need for greater efficiency, flexibility and reliability from generation assets.
The survey covered 250 professionals from utilities and technology companies across the Delhi-National Capital Region (NCR), Mumbai, Pune, Bengaluru, and Chennai. Nearly seven in 10 respondents said power generation offers the best opportunity for efficiency gains from AI over the next five years.
The survey also comes against the backdrop of rapid growth in India’s data centre infrastructure. Data centre capacity is projected to increase from 1.5 gigawatts (GW) in 2025 to 7 GW by 2030, while electricity demand from data centres alone is forecast to reach 13.56 GW by financial year 2031 to 2032.
Energy companies are also increasing their investments in AI- and renewable-energy-related infrastructure. Adani and Reliance have committed more than $210 billion to AI-driven infrastructure and renewable energy through 2035, according to ENCIS.
Predictive asset management and proactive maintenance emerged as the most promising applications of AI in power generation. About 54 per cent of respondents said these technologies could help power producers identify faults earlier, reduce outages and extend the operating life of expensive equipment.
Grid management and renewable energy integration were also identified as potential applications, but respondents saw the fastest and most significant benefits coming from the digitalisation of core plant operations.
The findings indicate that AI adoption in generation is being driven primarily by operational requirements rather than regulatory pressure. Efficiency was identified as the top reason for investing in AI by 61 per cent of respondents, while 22 per cent said their primary motivation was to accelerate the transition to clean energy. Only 9 per cent cited regulatory compliance as their main reason for investing in AI.
Despite strong expectations around AI, the survey found that power companies face significant challenges in deploying the technology.
More than 37 per cent of respondents identified a shortage of skilled professionals as the most significant barrier to AI adoption. A further 23 per cent cited difficulties in integrating new technologies with ageing infrastructure, while 18 per cent pointed to data privacy and security concerns.
ENCIS said these challenges are becoming more important as AI-driven electricity demand grows four times faster than in other sectors.
Abhishek Bhatnagar, Director General, ENCIS, said the survey showed that the sector recognised the potential of AI but was not yet fully prepared for large-scale deployment.
“The survey reveals a sector with conviction, but not yet complete readiness. Industry sees the value; the execution architecture, skills, data, legacy integration and governance must now catch up,” Bhatnagar said.
He said fragmented pilot projects would not create national advantage and called for policymakers, generating companies, technology companies, research institutions and capital providers to develop a common approach centred on trusted data, responsible AI and scalable deployment.
The objective, he said, should be to ensure that AI improves reliability, reduces costs, strengthens safety, accelerates decarbonisation and advances energy sovereignty.
Bhupinder Singh Bhalla, Chairman of Bharat Electricity Council at ENCIS and former secretary at the Union Ministry of New and Renewable Energy, said the next stage of power sector competition would increasingly depend on how effectively generation assets use digital technologies.
“By 2030, the decisive divide will not be between conventional and renewable generation; it will be between assets that learn and assets that merely operate,” Bhalla said.
India is adding generation capacity at historic scale, he said, but the next productivity gains would come from improving the availability, flexibility, safety and operating life of existing and new assets.
Bhalla said AI could enable these improvements when supported by trusted data, sound engineering, cyber-secure architecture and accountable human judgement.
“The objective is not autonomous power plants for their own sake; it is more reliable, affordable and resilient electricity for the economy,” he said.
Based on the survey findings and sectoral evidence, ENCIS identified five priorities for accelerating responsible AI adoption in India’s power generation sector.
The first is to focus on value-driven AI applications that directly improve operational outcomes including availability, efficiency, emissions and safety.
The second is to establish a trusted data framework across plants and fleets by standardising asset taxonomies, ensuring reliable sensor data and enabling secure and interoperable systems.
The third is to design AI systems specifically for critical infrastructure, with validation, explainability, cybersecurity and clear human oversight built into their deployment.
The fourth is to move beyond isolated pilot projects by scaling proven applications through standardised architectures and deployment protocols.
The fifth is to develop hybrid talent by equipping power sector engineers with data and AI skills while strengthening power system expertise among digital professionals.