From Pilots to Scalable Systems: How India Can Lead the Next Phase of Planet Scale AI
Bengaluru, 18 February 2026 — SatSure convened a high-level panel discussion titled “Planet Scale Intelligence for Economic Growth: AgriStack, ClimateAI and Foundational Models for Global Impact” at IndiaAI Summit 2026, bringing together leaders from infrastructure, enterprise transformation, development impact and industrial AI to examine what it truly takes to move artificial intelligence from experimentation to durable, planet-scale deployment.
Moderated by Prateep Basu, Co-founder and CEO at SatSure, the discussion was hosted by Divya Sharma, VP of AI/ ML at SatSure, with closing reflections delivered by Rashmit Singh Sukhmani, Co-founder and CTO at SatSure.
The panel explored a central paradox shaping the next decade of AI: while model capabilities are advancing rapidly, most deployments remain trapped in pilot phases, unable to translate into sustained operational systems. The conversation reframed AI not as a standalone technology challenge, but as an infrastructure problem requiring coordinated investment across energy, compute, sensing, institutional design, and human systems.
India was positioned as a growth area for this transition. Its population scale, digital public infrastructure, and real-world complexity expose both the limits of models and the opportunity to build intelligence as shared national capability.
Session Highlights
- From Models to Systems: The session highlighted that intelligence does not scale through larger models alone, but through aligned systems spanning energy, data centers, GPUs, applications, and institutional workflows.
- Affordability as Infrastructure Design: Speakers highlighted how industrial agents, and decision systems remain inaccessible unless the full stack is locally built and optimized for cost, reinforcing the need for sovereign AI capabilities.
- Human-Centered Adoption: One example highlighted how voice-first interfaces are essential in messaging apps, used extensively by blue collar workforce in India, where speech is more universal than typing, and thus, why technology must adapt to human behavior rather than expecting behavioral change.
- The POC-to-Production Gap: Most enterprise AI initiatives fail to scale because of frontline trust, incentive alignment, and learnability acting decisive factors.
- Multi-Layered Intelligence in Physical Systems: Sharing examples of drones, satellites, vision models, and regulatory knowledge graphs working together to monitor methane emissions, manage industrial risk, and enable compliance in energy infrastructure. Such, multi-layer intelligence systems are needed to solve real problems.
- India as a Global Sandbox: Panelist highlighted how innovations proven in India can travel globally, provided local ecosystems, data pipelines, and institutional systems are established.
- Enterprise Transformation at Scale: Drawing on ITC’s experience across agriculture, FMCG, and manufacturing, including satellite-based deforestation compliance, AI-driven supply chain resilience, hyper-local marketing, and quality automation.
Speakers
- Derick Jose – MD Industrial AI at Accenture
- Gaurav Aggarwal – Chief of AI at Reliance JIO
- Gaurav Kataria – CDIO, PSPD at ITC Limited
- Michael Tsan – Global Partner and AI Lead for Dalberg Advisors
Outcomes
The discussion underscored a decisive shift underway in artificial intelligence: from analytics toward operational intelligence, where AI systems participate directly in decision-making across infrastructure, supply chains, education, and climate-sensitive industries.
Participants converged on several actionable insights:
1. Treat Intelligence as Infrastructure
AI must be engineered as continuous operational capability, not episodic projects. This requires coordinated design across compute, energy, sensing, models, applications, and institutions.
2. Build for Human Reality
Mass adoption in India depends on voice-native, multilingual systems aligned with everyday behavior. Intelligence must meet people where they are, rather than imposing technical abstractions.
3. Close the Trust Gap in Enterprises
Scaling AI demands deliberate change management: aligning operator incentives, establishing empirical trust through benchmarking, and measuring system learnability rather than early-stage accuracy alone.
4. Integrate Physical and Digital Systems
High-impact applications increasingly combine satellites, drones, vision AI, and regulatory frameworks, enabling outcome-driven intelligence in sectors such as energy, agriculture, and manufacturing.
5. Move from Compute Capacity to Decision Economics
Infrastructure investments must ultimately translate into affordable decisions, whether in education, healthcare, climate risk, or industrial operations.
Architecting Intelligence as National Capability
Closing the session, Rashmit Singh Sukhmani emphasized the need to move beyond project-based deployments toward nationally coordinated intelligence systems spanning chips, compute, sensing, models, applications, energy, and governance. He highlighted that India’s strength lies not only in building technology, but in applying intelligence at scale, provided ecosystems collaborate rather than operate in silos.
For SatSure, the panel reaffirmed a core belief: the future of AI lies in operational systems that integrate Earth observation, physical infrastructure data, and institutional workflows into unified decision platforms.