India AI Impact Summit 2026 focuses on trust and accountability in AI

As artificial intelligence moves beyond experimentation into everyday systems, the debate is shifting from what the technology can do to how it should be governed. That was the central theme of a discussion on responsible and ethical AI at the India AI Impact Summit 2026, where policymakers, industry leaders, academics and infrastructure experts examined how AI can be deployed at population scale without compromising trust or accountability.

Participants included representatives from the Software Technology Parks of India (STPI), Mastercard, Kore.ai, Indian Institute of Technology Kanpur and the National Power Training Institute, reflecting the breadth of sectors now shaped by AI.

From capability to governability

Speakers noted that AI systems are increasingly influencing credit access, welfare distribution, healthcare triage, payments and public services. Once embedded in such high-impact areas, the primary concern becomes whether these systems can be trusted, validated and held accountable at scale.

A distinction was drawn between “responsible AI” and “ethical AI”. Responsible AI was framed in operational terms, including fairness, accountability, security, transparency and privacy. Ethical AI, by contrast, was described as a broader leadership consideration, encompassing environmental impact, social disruption and long-term economic consequences such as job displacement. The difference, participants argued, is significant: technical safeguards alone are insufficient without governance choices about when and how AI should be deployed.

Building trust at national scale

At India’s scale, governance was seen as inseparable from trust. Adoption of AI systems depends on public confidence, and without adoption, impact remains limited. Rather than aiming for zero risk, speakers suggested that systems should be designed for resilience, capable of failing safely, being audited transparently and corrected quickly.

The discussion also highlighted India’s ambition to move from being largely an AI user to becoming an AI creator. Domain expertise was described as a critical starting point. Instead of focusing solely on building large models, practitioners were encouraged to identify real-world gaps within their sectors and apply AI to solve practical problems.

Agriculture and ecosystem modelling

Agriculture was cited as an example of both opportunity and neglect. Despite its central importance, it receives a relatively small share of global AI investment. Early applications such as vertical farming have focused on optimising light, nutrients and climate control to increase yields. However, speakers argued that the next stage will require more complex ecosystem modelling, including pollination dynamics and microclimate simulation.

In this context, AI is being used not only to optimise plant growth but also to model biological processes in controlled environments. Sensor networks, airflow regulation, spectrum-specific lighting and climate systems can be integrated to create more stable and predictable agricultural ecosystems.

Rethinking academia and skills

From an academic perspective, participants questioned whether research output alone is sufficient to drive impact. They argued for a stronger “startup mindset” within universities, with greater emphasis on product development, adoption and real-world utility. One proposal suggested that advanced degrees could, in some cases, be awarded for commercially viable products rather than solely for published research.

A technical caution was also raised: AI systems are statistical in nature and can behave unpredictably. In high-stakes sectors such as healthcare and transport, human oversight remains essential. For India specifically, AI solutions were described as needing to be frugal and practical in order to achieve widespread adoption.

Critical infrastructure and cyber resilience

In the power sector, AI’s role was linked to future expansion and modernisation. With India aiming to significantly increase its generation capacity over the coming decades, technology will need to be embedded across the value chain. At the same time, cyber resilience was emphasised as critical, given the strategic importance of energy systems.

Bridging the skills gap between academia and industry was identified as another priority. Participants suggested that AI competencies, along with ethics and responsibility, should become core components of engineering education rather than optional subjects.

An ecosystem approach

The closing discussion framed India’s AI moment as an ecosystem challenge rather than a race to build the largest model. Government initiatives were described as supporting multiple layers of development, from compute infrastructure to applications and foundational models. Expanding access to computing resources was presented as a way to broaden participation in AI development.

However, speakers emphasised that technical capability alone does not guarantee success. Validation, oversight and public trust remain essential. If AI systems can be deployed responsibly across diverse and complex environments, India may offer a practical model for other countries facing similar constraints.

The overarching message was clear: long-term advantage in AI will depend less on scale of models and more on the ability to deploy systems fairly, transparently and securely in sectors where failures carry real-world consequences. In that sense, trust was presented not as an accessory to innovation, but as its foundation.