Inc42 AI Summit 2026: India's $126 Bn AI Dream Hits Reality

Key Takeaways

• India remains a consumer of global AI infrastructure, not a builder, according to InMobi's Mohit Saxena
• Rapido now generates two-thirds of its code using AI, compressing development cycles to one week
• Investors are moving past AI buzzwords to scrutinize adoption rates, margins, and compute costs

The Paradox at the Center of Indian AI

More than 600 startup founders and CXOs gathered in Bengaluru for the Inc42 AI Summit 2026. The mood was ambitious. India's AI market is projected to hit $126 billion by 2030. But the conversations on stage told a different story.

Speaker after speaker pointed to the same problem: India is building AI applications on top of infrastructure it doesn't control. The country lacks domestic chip manufacturing. Venture capital for long-horizon AI bets remains thin. And the ecosystem patience required to build foundational technology is in short supply.

“India is still mostly a consumer of global tech rather than a builder of core infrastructure.”

— Mohit Saxena, Cofounder & CTO at InMobi

Saxena's assessment set the tone for the summit. His argument was blunt: India is not in the same league as the US or China when it comes to AI fundamentals. The country can build impressive applications, but it depends on foreign chips, foreign cloud providers, and foreign foundation models.

Distribution Beats Technology

Rahul Chari, cofounder and CTO of PhonePe, offered a different lens. He argued that AI's lower entry barriers make distribution and data more valuable than ever. When anyone can spin up an LLM-powered feature, the competitive moat shifts to who has the users and the data to train on.

“The focus must shift from high-level theory to the hard parts of production: managing rupee economics, building for 22+ languages, and ensuring reliability.”

— Rahul Chari, Cofounder & CTO at PhonePe

Chari predicted that engineers will evolve from coders into systems architects. Writing code will matter less. Understanding hardware constraints, system design, and user behavior will matter more.

Inside Rapido's AI Factory

The most concrete example of AI adoption came from Rapido. The ride-hailing company revealed that two-thirds of its code is now AI-generated. Software production cycles that once took months now finish in a single week.

This isn't a pilot program. Rapido has restructured its engineering organization around AI-assisted development. Human engineers focus on architecture decisions and edge cases. The AI handles boilerplate, testing, and routine implementation.

Fintech platform S45 took an even more aggressive stance. The company only hires humans when AI cannot do the job. Every role starts with the assumption that AI can handle it. Human hiring happens only after that assumption fails.

Voice and Velocity

Zepto and ElevenLabs showcased two different applications of the same technology shift. Zepto uses AI to accelerate experimentation. The quick-commerce company can now test hypotheses faster, iterate on features faster, and ship faster.

ElevenLabs focuses on the customer side. The voice AI company helps brands convert voice interactions and user intent into reduced friction and better resolution. When a customer speaks, the AI understands what they want and routes them to the right solution.

What Investors Actually Want to See

The funding conversation at the summit was revealing. Archana Jahagirdar of Rukam Capital and Ashwin Raguraman of Bharat Innovation Fund said investors have moved past AI buzzwords. The questions they ask now are harder.

• What is actual customer adoption, not demo signups?
• What are real margins after compute costs?
• How much does it cost to run inference at scale?
• Is there real customer usage, or just hype-driven trial?

The shift reflects a maturing market. In 2024 and 2025, AI was a magic word that opened checkbooks. In 2026, investors want proof that AI companies can build durable businesses, not just impressive demos.

The 4-8 Week Playbook

For founders looking to sell AI solutions to enterprises, the summit offered a practical timeline. Speakers recommended a 4-8 week pilot playbook to clear enterprise security and legal gates. This is the realistic timeline for getting past procurement, compliance, and IT security reviews.

The advice reflects hard-won experience. Enterprise sales for AI products stall not because of technology issues, but because of legal and security concerns. Founders who plan for this timeline close deals. Those who expect faster cycles get stuck in limbo.

The Gig Economy's Parallel Crisis

The summit also touched on a related issue: safety gaps in India's gig economy. Multiple complaints involving gig workers have surfaced online in recent months, triggering outrage and raising concerns about user safety.

Platforms remove offending workers when incidents come to light. But gig worker unions argue that platforms cannot escape accountability for gaps in verification, onboarding, and enforcement. The debate is whether platforms are technology providers or employers with safety obligations.

What Comes Next

The biggest takeaway from the summit was a shift in framing. The question is no longer whether AI matters. Everyone agrees it does. The question is which teams can turn AI leverage into durable business advantage.

For Indian startups, this means competing on execution rather than technology access. The same foundation models are available to everyone. The winners will be those who understand Indian users, manage costs in rupees, and ship production-quality products.

Frequently Asked Questions

What is the projected size of India's AI market by 2030?

India's AI market is projected to reach $126 billion by 2030, according to figures cited at the Inc42 AI Summit 2026.

How much of Rapido's code is AI-generated?

Rapido revealed at the summit that two-thirds of its code is now AI-generated, compressing software production cycles to a single week.

What are investors looking for in AI startups in 2026?

Investors are asking tougher questions about actual adoption rates, margins after compute costs, and real customer usage rather than accepting AI buzzwords at face value.

How long does it take to clear enterprise security gates for AI products?

Summit speakers recommended a 4-8 week pilot playbook as the realistic timeline for clearing enterprise security and legal gates.

Why is India considered behind the US and China in AI?

According to InMobi's Mohit Saxena, India lacks the same depth of capital, chip access, and long-term ecosystem patience, making it a consumer of global tech rather than a builder of core infrastructure.

Source: Inc42 Media / Team Inc42