Videos
Physical AI & the Future of Industrial Operations
Insights
- Physical AI is moving automation beyond rigid industrial systems toward adaptive, context-aware operational environments.
- The next phase of industrial transformation will depend on combining AI-driven systems with human expertise and decision-making.
- Enterprises that integrate AI directly into operational workflows will unlock greater resilience, productivity, and responsiveness.
Deepak Subramanian, Head of Solutions & Strategy, Insurance at Infosys explores how physical AI is beginning to redefine industrial operations by bringing intelligence, adaptability, and real-time decision-making into physical environments. He explains how industries are moving beyond traditional automation toward systems that can interact dynamically with people, machines, and changing operational conditions. The discussion highlights the growing importance of combining AI-driven automation with human judgment rather than viewing AI purely as a replacement for labor. He also examines how enterprises must rethink workflows, workforce models, and operational design as AI becomes increasingly embedded into industrial processes. Looking ahead, he outlines how organizations that successfully connect physical AI with operational execution will be better positioned to scale productivity, resilience, and long-term transformation.
Deepak Subbramaniam:
One of the biggest drivers in life and annuity subdomain is about the push for AI in core workflows. The reason for a push for AI is because there's a lot of expense pressures. The expense ratios for life and annuity industry are very thin as such. So they need to find ways to do things in a more efficient manner. The second driver is customer expectations. The customer expectations are changing because of the experiences that they're having from all the other different engagements they're having with the industry.
Life and annuity has a lot of documents, a lot of different complex process flows. So one of the clear evidences that it is ripe for automation, it is ripe for AI. The reason being is lot of reasoning happens in life and annuity space. And wherever there is reasoning that is happening, you see a potential of AI.
I feel that AI is proven, but not yet fully transformational as such. The insurance industry is still grappling with what are the right use cases, which will give them the needed ROI for this.
Creating agents is not that difficult. But integrating it with the ecosystem to get the desired business outcome and be able to achieve that ROI is something that they are still figuring it out. And the reason, one of the major reasons why they feel that is because of insurance being a very regulated industry. They need to explain every decision to the audit committees. It's not just about getting the output. It's about how they've got the output.
One of the other things that we have started to realize in many of these companies is there is a pilot fatigue is what we call. There's so many pilots that are happening without seeing the end of the tunnel in terms of the outcome. And one of the biggest reasons is the pilot does well, but when it actually is taken to the larger ecosystem, it fails to realize the value.
You need to really rethink about the operating model as such, how human and agents are going to come together. Because we know for sure humans and agents are going to be collaborating together in future. So it's very important for humans to understand how they're going to use agents in their work. It's more about amplifying the human potential rather than actually bringing agents and taking their work as such.
One of the biggest opportunities that I see for life and annuities clients is to reimagine their current business processes as such. Today, traditional business processes that they have is very human intensive. I think it's a great opportunity for them to really look at agent-human collaborative ecosystems, where they can significantly drive the business outcome that they are trying to achieve.
