BENGALURU: Artificial intelligence systems are improving at high speed, while businesses and governments are struggling to absorb them, both Dario Amodei, chief executive officer of Anthropic, and Nandan Nilekani, chairman of Infosys, said at the India AI summit.
Amodei said AI models are becoming highly capable, particularly in software engineering and biomedical research. “We’re getting models that are very good at software engineering that are increasingly good at biomedical innovation,” he said.
But stressed that capability and real-world impact are not the same. “There is this duality between the fundamental capabilities of the technology and the time that it takes for those capabilities to diffuse into the world,” he said.
Even if AI progress stopped today, he said, companies could generate far more economic value than they currently do. The barrier is not intelligence, but adoption. “There are just frictions to adopt things through enterprises,” he said.
In simple terms, AI may be racing ahead in laboratories, but it moves more slowly through offices, factories and public systems.
The risk of backlash
Nilekani warned that if AI does not deliver visible public benefit, it could face political resistance.
“The anger of blue-collar workers led to the collapse of globalisation,” he said. “The anger of white-collar workers is going to lead to the collapse of AI.”
He argued that people must see practical gains in farming incomes, healthcare, education and access to services. Otherwise, frustration over job losses, misinformation and rising costs could grow.
AI, he said, must work “for a billion people”, not just for technology firms.
India as a testing ground
Both men suggested that India could play a central role in this next phase of AI.
Nilekani pointed to India’s experience in rolling out digital public infrastructure at scale, including identity and payments systems that serve more than a billion people. He said diffusion is “both an art and a science” involving institutions, policymaking, trust-building and negotiation.
India, he said, should aim to become “the use case capital of the world”.
Amodei said there is a strong technical enthusiasm among Indian developers and high usage of cloud tools for programming and mathematical tasks. “There’s just excitement here and technical acumen,” he said.
He also said that AI could greatly increase economic growth in countries that have strong technical talent and room to grow. In rich countries, he has wondered if AI could raise growth to around 10%. In India, he said growth of 20 to 25% may sound “absurd”, but it could be possible if the right conditions are in place.