Krutrim, the AI arm of Ola, has repositioned itself as a focused domestic AI Cloud Services provider after shelving its chip design ambitions, betting on faster scale and profitability even as analysts debate whether the move marks strategic focus or a lateral shift within India’s crowded GPU infrastructure market. Experts say that frequent pivots like these can raise eyebrows, especially if earlier ambitions were positioned strongly.
This move follows a business realignment in late 2025, which involved reallocating capital and talent, including a pause on chip design initiatives to concentrate the company’s resources on building and scaling its core AI cloud services stack. The result is a full-stack cloud service built in-house and deployed at scale without external dependencies.
Revenue growth and profitability mark key milestone
The company reported revenues of around ₹300 crore in FY26, a threefold increase over FY25, and its first annual net profit, with a profit after tax margin of over 10%. Krutrim is now financially self-sustaining, with no immediate requirement for external funding, including from its founder.
Alongside, Krutrim shared that it is seeing increasing adoption with over 25 large enterprise customers, including telecom service providers, top financial institutions, consumer internet platforms, AI and deep-tech companies, healthcare, logistics platforms, and digital-first enterprises, a cross-sector roster that reflects the platform’s applicability beyond the Ola Group. A majority of Krutrim’s GPU compute capacity is committed to external enterprise workloads.
Experts see strategic refocus but flag execution challenges
This pivot, according to Manish Mohta, Founder, Learning Spiral, may be due to the practicalities associated with chip design, which require substantial funding and time to develop.
“Typically, global giants hold the vast majority of expertise in chip design. A newer company entering this environment would find it difficult to maintain competitiveness from a hardware perspective. The pivot to prioritize speed-to-market and scalability vs long-term investments implies a strategic decision to focus on investment areas that offer the possibility of differentiation or ROI more efficiently in shorter timeframes (short to mid-term).”
“Krutrim’s decision to pause its chip programme and post its first annual profit as a domestic AI cloud provider appears to be the company choosing the more achievable path. Examined through the five layers that constitute the AI stack, namely energy, chips, infrastructure, models, and applications, one can see that it moved from layer two to layer three. It is still entirely within the supply side of the stack,” Nikhar Arora, Director and Builder, BOTS.AI by HR Anexi, echoed.
Krutrim’s arc, he said, is not a story of failed ambition nor the strategic discipline that its announcement framed it as. The company, launched in 2023, promising to build India’s complete AI stack, has now arrived at the same layer where every other GPU cloud vendor in India is competing. The hardware economics made the chip decision rational. But moving from layer two to layer three is not a pivot into differentiated territory. It is a lateral move within a conversation India has been having with itself since 2023.
“This looks more like a strategic refocus than a retreat. In fast-evolving sectors like AI, it’s common for companies to reassess priorities and double down on what’s working. The key is whether the new direction is sharper and more executable,” Jaspreet Bindra, Co-founder & CEO, AI & Beyond, said.
He added that the AI cloud is a tough space, given the dominance of global hyperscalers. However, differentiation could come from focusing on India-specific use cases, language models, and local enterprise needs. If Krutrim can build solutions tailored to the Indian ecosystem, it can carve out a niche rather than competing head-on with global giants.
Moreover, scaling a reliable AI cloud platform is not easy, especially without a long history in infrastructure. Success will depend on execution, partnerships, and how quickly they can build credibility with developers and enterprises. AI infrastructure and cloud are still highly capital-intensive, even without chip design. It will require sustained investment.
Mohta added that creating an AI cloud platform from the ground up requires significant engineering, infrastructure investment, and operational excellence -- challenges that are particularly pronounced when building an AI Cloud with limited experience, as Krutim has. Execution will be challenging, but hiring, partnerships, and leveraging existing cloud layers can enable scale. Ultimately, success hinges on uptime, speed, and developer adoption—without consistent performance and trust, scaling will falter despite the market opportunity.
“A company’s constant change of direction can be viewed as a lack of stability and will likely generate doubt surrounding its capacity to fulfill its obligations, especially if it made aggressive promises. Consistent changes in direction may lead to doubt about its long-term objectives among investors and partners. The most substantial risk is over-promising without meeting expectations. To sustain or increase investors’ and partners’ confidence in the broader vision, transparent communication of expectations, realistic milestones, and developing a track record of consistent execution will be critical,” he said.
The company did not respond to a detailed questionnaire shared by businessline by the time of going to press.
Published on May 6, 2026