With AI, challenge is convincing people that learning still matters: Coursera CTO

At a time when artificial intelligence (AI) has made information easier to access than ever, the challenge confronting education platforms is convincing people that learning still matters. For Coursera, the New York-listed learning company with a market valuation of about $1.5 billion, the answer lies not in more content but in helping people translate information into real-world skills, said its chief technology officer Mustafa Furniturewala.

“It’s not possible to put the people who require learning in front of a chatbot and expect them to learn,” Furniturewala told ET in an exclusive interview. “The grounding from the expert content as well as the pedagogy that’s behind learning, is still important.”

Yet, he believes the industry’s growing obsession with AI models and benchmarks is missing a larger opportunity.


“I believe there is a lot of innovation that can happen on the application layer,” he said. “The value generation is more about what’s built on top of the frontier models with the ecosystem you create and with the application layer you create. The next stage will be how you actually create an ecosystem that generates value with AI.”

As companies worldwide continue to invest aggressively in AI while struggling to demonstrate clear returns, Furniturewala said the real bottleneck is no longer access to AI models but whether organisations have the skills to use them effectively.

“There is a lot of upskilling that’s required,” he said. “How do you actually give the AI the right context? How do you connect it with the right tools? That’s why, how you use the tool is extremely important.”

India is emerging as one of the most important markets in that transition. Coursera now has more than 35.4 million learners in the country, making it the company’s second-largest market globally. India also leads the world in generative AI learning on the platform, with more than 4 million GenAI enrolments, while enterprise GenAI enrolments have increased 179% year-on-year.

Furniturewala believes the country has an opportunity to create significant value even if it is not currently home to a frontier AI model.

“India has such a big talent pool of young people that are the perfect age group to upskill in AI and leverage this technology,” he said. “Whether or not there will be a foundation model in India, it might have something to do with the need for it. But I expect to see a lot of innovation in India happening around AI.”

The debate over AI’s impact on jobs has also evolved significantly over the past year. While early fears were largely on large-scale job displacement, Furniturewala argued that the strongest outcomes are emerging when AI is paired with human expertise rather than used as a replacement for it.

He pointed to Coursera’s own development process as an example. More than 90% of the code behind Ollie, the company’s upcoming AI-powered microlearning product, was generated using AI.

“But could we have done it without very senior engineers? Absolutely not,” he said. “The people behind that were actually senior engineers who were using AI in a thoughtful way, with the right tools, the right context and the right architecture.”

This philosophy is also shaping Coursera’s approach to the future of learning. As platforms compete for attention in an increasingly crowded digital environment, the company is preparing to launch Ollie, a mobile-first microlearning app developed using Claude Code, and built around 90-second lessons, AI-powered conversations and personalised learning journeys.

“We are spending our time on screens and doing these doom-scrolling activities,” Furniturewala said. “Ollie is meant to divert time towards something that’s valuable. Instead of spending time doom scrolling, I am growing my skills as a person and doing something valuable for me.”

The future of learning, according to him, will not be determined by who has access to information given that in the AI era, information is increasingly becoming a commodity. The bigger challenge, he said, is helping people turn that information into capability.