Past AI hype: Why knowledge is the true aggressive moat

As organizations race to undertake AI, a quieter however extra decisive shift is below means: knowledge has emerged as the true supply of aggressive benefit. In a Snowflake panel dialogue titled ‘Future-ready Startups: Information, Product Innovation & Capital Technique for 2026’, leaders throughout fintech, healthcare, investing, and cloud infrastructure unpacked what it takes to show fragmented, delicate, and infrequently imperfect knowledge into methods that scale with belief. 

Panel dialogue contributors included Govindarajan Kadambi, Co-founder and CPTO, Dvara Holdings; Aravind Ganesan, CTO, Kauvery Hospitals; Guruswamy Ramasubramanian, Normal Associate at Sirius One Capital; and Sunith Chacko Philip, Supervisor – Industrial & Digital Natives (India), Snowflake. The panel was moderated by Shivani Muthanna, Senior Director, Content material Partnerships, YourStory.

The state of information: challenges, alternatives and belief

Kadambi opened the dialogue by explaining how Dvara Holdings serves neglected and underserved prospects with merchandise together with financial savings, investments, and earnings tax submitting. He recognized the first knowledge challenges as buying and utilizing knowledge successfully to prioritize key components similar to affordability, reliability, and belief.

Affordability evaluates if prospects can afford companies long-term for lifetime worth and cross-selling. Dvara assesses this by way of earnings benchmarks and private components like spending patterns, household dimension, and age. Reliability focuses on service consistency, whereas belief is crucial—particularly for a phase exploited and skeptical of digital instruments and unknown entities.

Commonplace fintech fashions falter on account of earnings unpredictability. “For us, fintech or banking fashions do not likely work as earnings knowledge is irregular. Take fruit distributors or cab drivers, we have to perceive their stream of earnings. Some months, bills match their incomes; in different months, their bills exceed earnings. Numerous the problem has been round getting this knowledge and becoming it into affordability,” Kadambi mentioned.  

He additionally famous that the place totally digital options fail regardless of value advantages, phygital approaches succeed. Incorporating bodily touchpoints like in-person interactions, cellphone availability, or partnerships with trusted self-help teams is crucial when dealing with hard-earned cash.

Ganesan, of Kauvery Hospitals, distinguished healthcare knowledge into two realms: extremely delicate scientific affected person data and customary administrative knowledge, generated in a different way. Most scientific knowledge stays non-digital, as medical doctors typically use paper information, even throughout hospital admissions. Kauvery Hospitals has prioritized digitizing this delicate knowledge amid regulatory and hospital-driven pushes. Now superior on this journey, the hospital views scientific knowledge – not administrative – as a strategic asset for distinctive affected person insights and mental property.

Over the previous couple of years, Kauvery constructed a knowledge platform and digitization ecosystem, now harnessed right into a lakehouse or knowledge lake structure, which lets the group establish related AI use instances. “AI has commoditized intelligence. There is no such thing as a strategic worth in that anymore. Though we have now an investing arm targeted on AI, we’re now not differentiating primarily based on simply AI capabilities. We take a look at how intelligently startups are utilizing knowledge. Information is the strategic asset,” Ganesan mentioned. 

Ramasubramanian, of Sirius One Capital, emphasised that knowledge lies on the coronary heart of innovation—what organizations make of it issues most. Present startups maintain a slight edge, having collected huge datasets throughout retail, ecommerce, marketplaces, and past. He famous that beforehand they lacked instruments to derive intelligence or monetize this knowledge successfully. Now, with superior AI capabilities, they’ve a major alternative to unlock its worth.

Information foundations: the infrastructure of intelligence 

“There’s no AI technique, with out a knowledge technique,” mentioned Sunith Chacko Philip of Snowflake. He mentioned evolving knowledge foundations within the AI period. Not like previous efforts to interrupt inside silos, as we speak’s AI revolution necessitates a sturdy knowledge technique because the prerequisite for any AI initiative.

In line with Philip, startups typically launch minimal viable merchandise (MVPs) to resolve particular issues, attaining speedy adoption and “escape velocity” by way of the hockey-stick development sample. Nonetheless, that is the stage the place backend infrastructure typically turns into brittle, inflicting breakages, assist overload, and score dips. Snowflake permits startups to construct purposes on a scalable core knowledge basis, whether or not it’s from MBs or GBs of information. This eliminates infrastructure administration points throughout hypergrowth, permitting give attention to function rollout.

Philip cited a latest instance: A retail intelligence platform, akin to produce chain optimization for kirana shops, automated handbook stock monitoring that was beforehand dealt with by means of paper and Excel sheets. Nonetheless post-launch, knowledge volumes exploded, overwhelming the sluggish legacy methods. By leveraging Snowflake as a knowledge basis, the platform can now deal with mammoth workflows, rushing up function growth and accelerating market rollout quicker. “What we evangelize again and again is to have a really sturdy knowledge basis that primarily turns into an enabler for your corporation. AI can then develop into a catalyst to assist startups transfer ahead,” he mentioned.

Ganesan concurred, describing Kauvery Hospitals’ purpose of pursuing a unified “golden journey” for sufferers throughout digital and bodily interfaces—from app/WhatsApp bookings and AI-transcribed physician visits to ICU IoT displays, scans, and pictures. This generates various knowledge modalities, which require a sturdy knowledge basis.

He shared that the group is constructing a knowledge lakehouse structure to supply, consolidate, sanitize, and standardize knowledge. Whereas rules proceed to evolve in India, this setup will unlock knowledge’s full potential.

A human within the loop: The guardrail of agentic AI

The dialog turned to agentic AI as a key differentiator for AI-native startups and panellists agreed with the thought whereas advocating for a cautious method. 

Ganesan careworn that healthcare stays a “human-in-the-loop” ecosystem, even with agentic AI. As a consequence of agentic AI’s non-deterministic nature, rules prohibit granting AI full company relating to scientific choices. In line with Ganesan, this can be a barrier that’s unlikely to be lifted quickly. Kauvery Hospitals has deployed agentic AI on the non-clinical facet, utilizing it to streamline administrative workflows similar to buy orders, bill processing and straightforward automation. One other profitable use case entails utilizing AI to transcribe doctor-patient interactions, releasing up medical doctors from typing or sustaining eye-contact with sufferers. Medical doctors can evaluation and approve transcripts, enhancing effectivity with out changing oversight.

Ramasubramanian added that latest portfolio additions leverage the AI and agentic wave, primarily automating end-to-end enterprise workflows. He concurred with Ganesan on the “human-in-the-loop” mannequin, sharing that crucial choices all the time require human oversight versus totally autonomous actions.

The dialogue lined a variety of matters, from knowledge pipelines to funding standards and strategic shifts mandatory for 2026.

Philip highlighted investor confidence in Snowflake-backed startups and their tech stacks, and revealed how the group reached out to enterprise capitalists to put money into seed/pre-seed ventures to construct MVPs immediately on the platform.

Kadambi defined that regardless of customary unit economics for mature merchandise like financial institution accounts, investments, and insurance coverage, serving underserved segments requires Dvara to adapt to distinctive conditions. When fully-digital fashions failed, the corporate needed to pivot, re-architecturing right into a customer-facing, agent-assisted group.

Lastly, Ramasubramanian suggested healthcare startups on key hurdles. Gross sales cycles drag for six–8 months, not weeks; whereas high quality knowledge/IP stays locked in establishments amid uncertainties round sharing personally identifiable data (PII) and delicate private data (SPI). To construct a very good healthcare product, startups should accomplice with establishments and medical doctors, and onboard clinicians by providing fairness, as they are going to be essential in validating a product and discovering product-market match.