07 February 2025 : The Hindu Editorial Analysis

1. The saga of regulating India’s thermal power emissions

(Source – The Hindu, International Edition – Page No. – 8)

Deadline Extension Without Justification

• This delay affects around 20 GW of thermal plants, which are located in densely populated areas.
• No reason was provided for the extension, marking another delay in a decade-long process of implementing emission norms.

Revised Norms Introduced in 2015

• In December 2015, MoEFCC introduced stricter emission norms after public consultations.
• All thermal plants were expected to comply by December 2017.
• The norms included limits on particulate matter emissions and, for the first time, introduced standards for SO₂ emissions.
• These norms aligned with those in countries like Australia, China, and the United States.

Shifting Focus in the Debate

• Indian coal has lower sulphur content, making it easier to meet SO₂ emission norms compared to high-sulphur coal.
• However, discussions focused on challenges in using Flue Gas Desulphurisation (FGD) technology, which removes sulphur from high-sulphur coal.
• FGDs were never mandatory, but debates centered around their high cost, long installation time, and supply chain issues.

Government Reports and Changing Opinions

• 2020-2021: The Central Electricity Authority (CEA) questioned uniform emission norms and proposed extending deadlines to 2035.
• 2022: A study by IIT Delhi found FGDs improve air quality but recommended delays due to high costs and increased greenhouse gas emissions.
• 2024: A study by NITI Aayog and CSIR-NEERI suggested SO₂ norms were less important than particulate matter norms for air quality.
• Despite these studies, a consensus was never reached before norms were first implemented, leading to repeated extensions.

Repeated Extensions and Varying Deadlines

• Over the years, MoEFCC weakened some norms and extended deadlines four times.
• New deadlines:
  • Particulate matter emissions: Final deadline December 31, 2024 (some plants had to comply by 2022-2023).
  • SO₂ emissions: New deadline December 31, 2027.
• There is no public data on whether plants follow particulate matter norms.

Financial Burden on Electricity Consumers

• Many plants tendered contracts for FGDs but not fast enough to meet deadlines.
• Electricity regulators allowed plants to pass FGD costs to consumers, even if they do not meet emission norms.
• Current status:
  • 22 GW of thermal plants have installed FGDs.
  • 102 GW (nearly 50% of India’s thermal power) is in advanced FGD installation stages.
• With deadlines extended, many plants may not use FGDs to save costs, leading to:
  • Consumers paying for unused equipment.
  • No air quality improvement for nearby areas for at least three more years.

Long-Term Implications

• Delays in SO₂ norms cause serious environmental, health, and financial impacts.
• It is uncertain whether India will learn from these delays and improve pollution control policies.

Conclusion

• Repeated extensions in SO₂ emission norms show regulatory inefficiencies and financial strain on consumers.
• These delays increase environmental risks and raise concerns about India’s commitment to pollution control and public health.

2. Should India build a sovereign, foundational AI model?

(Source – The Hindu, International Edition – Page No. – 9)

Why Should India Build an AI Model?

• India should develop foundational AI models to build expertise and technological skills.
• It is important to have people who can create and improve AI models and develop applications on top of them.
• Sanctions on AI-related technologies like chips and software could impact India’s AI growth.
• Open-source AI models exist, allowing India to modify and use them without starting from scratch.

Is Sovereignty a Major Concern?

• Some believe sovereignty should be a reason for building an AI model, but others argue that India can adapt open-source models instead.
• AI development depends on advanced chips, which India does not currently manufacture.
• If global restrictions are imposed on AI technology, India could still use open-source AI models.

Financial Challenges of Developing an AI Model

• Creating a high-quality AI model is very expensive and requires hundreds of millions of dollars.
• Even a low-cost AI model requires millions for training and ongoing costs like salaries and infrastructure.
• India’s AI market is smaller than that of other countries, making it harder to recover the investment.
• Businesses in India mostly serve global markets, not just the local market.

Government’s Role in AI Development

• The government has announced efforts to provide AI resources, such as GPU clusters, at lower costs.
• This move can help startups and researchers by making AI training more affordable.
• However, big tech companies invest billions in AI, while India’s resources are more limited.

Where Should India Focus Its AI Efforts?

• Instead of competing directly with the world’s largest AI models, India should focus on targeted AI solutions.
• AI should be developed for important needs like Indian language processing and speech recognition.
• AI models should be designed for local use cases where they can make a big impact.
• The key question is how to improve India’s research environment and attract private investments.

Conclusion

• India should wisely allocate its resources and focus on practical AI applications rather than competing with global giants.
• Developing an AI ecosystem with strong research, investment, and innovation is more important than just building one AI model.