Original Coverage & Source Attribution: www.thehindubusinessline.com
India’s AI-native startup ecosystem is entering a faster growth phase, driven by falling AI model costs, improving infrastructure and wider access to AI capabilities, AWS said.
| Photo Credit:
ISABEL INFANTES
India’s AI-native startup ecosystem is entering a phase of faster growth as falling model costs, improving infrastructure, and wider access to AI capabilities allow young companies to move more rapidly from experimentation to commercial deployment, said Jason Bennett, vice president and global head of startups and venture capital at Amazon Web Services (AWS).
“The economics are changing much faster for startup delivery today than we did five years ago,” Bennett said.
Bennett noted that ‘speed and scale’ with which startups can onboard customers is changing rapidly as the economics of building AI products change, with token costs falling across both frontier and open-weight models.
This is prompting startups to adopt a more flexible approach to AI infrastructure. While companies may initially use frontier models, Bennett said many are increasingly evaluating open-weight models to manage costs and improve control over their technology stack.
Indian startups deploy AI at scale
The shift is visible across Indian startups. Pramaana Labs, an AI company focused on regulated workflows across tax, law, insurance and compliance, uses AWS infrastructure to fine-tune its 27-billion-parameter prover on an eight-node NVIDIA B200 GPU cluster. DevRev uses Amazon Bedrock to build AI that resolves routine enterprise customer issues, while Locobuzz has used Amazon Bedrock and Amazon Nova models to cut response latency by 30% and reduce costs by 50%.
Ultrahuman, meanwhile, uses Bedrock for AI-powered customer support and health-data applications. Its AI customer-support system has reduced time to first response by 75%, while median ticket resolution time has fallen 84%.
Global expansion opens new opportunities
Bennett said Indian startups are also increasingly looking beyond the domestic market, and their experience building for India can become an advantage as they expand overseas.
“The most valuable thing for many of those startups is to have a partner that can give them some insight about how they go global,” he said.
AWS estimates that AI-native startups in India are growing faster than their global peers. Research Strand Partners conducted for AWS found that Indian AI-native startups reported average annual revenue growth of 167%, compared with 156% globally and 74% for Indian startups overall. The research also found that 75% of Indian AI-native startups are building proprietary AI capabilities.
Enterprise AI and physical AI next
The next opportunity could lie beyond consumer-facing generative AI applications. Bennett said enterprise transformation remains relatively nascent, with opportunities across areas such as operations, finance and logistics. He also pointed to physical AI and world models as emerging areas.
“Enterprise transformation is still, I would say, very nascent,” Bennett said.
AWS expands startup and AI infrastructure support
AWS is also expanding its support for early-stage companies through AWS Startup Advisor, an AI-powered assistant that guides architecture, infrastructure, security and costs. The company said AWS Activate has provided more than $8 billion in credits to 350,000 startups globally.
AWS is also investing heavily in India’s cloud and AI infrastructure. Amazon has committed more than $21 billion between 2026 and 2030 towards AI and cloud infrastructure in the country, including an additional $13 billion announced during Amazon CEO Andy Jassy’s recent India visit.
