(This is part two of this interview. If you missed part one, you may read it here.)
Chirag Agrawal serves as Global Head of Data Science at Novelis. Novelis Inc. is a global leader in aluminum production and recycling, known for being the largest producer of flat-rolled aluminum products and the world's largest recycler of aluminum. Headquartered in Atlanta, Georgia, Novelis operates 33 manufacturing locations in nine countries and employs more than 14,000 people worldwide.
At Novelis Chirag has spearheaded transformative technology initiatives since 2019. He previously worked as Data Science Manager at The Kini Group (2016-2018) and held strategic roles at CNH Industrial and Ernst & Young.?
He holds a Master's degree in Machine Learning from Harrisburg University of Science and Technology (2017-2019) and a B.Tech in Mechanical Engineering from IIT Dhanbad. His expertise spans AI/ML, cloud solutions, Industry 4.0, and sustainable manufacturing
I had the opportunity to interview Chirag Agrawal about AIoT and his views on how AI is impacting Industry 4.0 and Industrial IoT. Here is the second half of the interview. If you missed the first part, you may read it here.
Carl Ford (News - Alert) (News - Alert): Another keen interest you have is regarding sustainability. Is there a specific area you want to call attention to today?
Chirag Agrawal: The summer holidays of my childhood were spent in my hometown, where water was always a scarce commodity. It is etched in my mind, seeing my father and mother sitting with us children, demonstrating how to capture rainwater in large barrels and explaining how to use every single droplet, how to save every drop. It was a life teaching; there is a limit as to how much one can spend, and there is always a cost of wastage. It struck me, in retrospect, when I saw the amount of water and electricity used in large factories, that if we can utilize technology to address these issues on an epic scale, then we can actually make a meaningful difference. Today we see no common mainstream KPI even within an ecosystem of partners. The lack of agreed-to KPIs for reporting was not sustainable.
AI and IoT can transition sustainability from a reactive to a proactive mindset for energy optimization, water management, and minimizing waste. The combination of AI and IoT devices allows for closed-loop, real-time management of the "sense to decide to act" process chain, for continuous improvement in sustainability KPIs.
CF: Lately the hyperscalers, such as Amazon and Google (News - Alert), have put a lot of effort into security, and more energy, which they are buying from nuclear power plants. If they are concerned about these issues, what does that mean for industry?
CA (News - Alert) (News - Alert): On energy, the big four (Amazon, Google, Microsoft (News - Alert), Meta) have shifted from “any renewable” to the more demanding expectation of firm, carbon-free power available 24/7 because the load from AI and data centers is growing explosively, and intermittent sources won’t cut it. That’s why you see Microsoft (News - Alert) with Constellation in a PPA for nuclear with Three Mile Island, Google with Kairos Power SMRs, and Amazon’s 1.92 GW deal tied to Susquehanna and SMR development rights. The portfolio shift is about reliable and scalable decarbonization. Multiple estimates suggest that data center power could more than double by 2030, with substantial contributions from nuclear, gas, storage, hydro, and geothermal sources.
On security, the hyperscalers continue to invest in zero-trust, confidential computing, supply-chain attestation, hardware root-of-trust, and continuous monitoring. The sensitive data-to-attack surface problem is only exacerbated by AI (model weights, embeddings, prompts, tools, agents, APIs). Regardless of the differences, the pattern remains congruent with industrial norms, such as the CPwE/IEC (News - Alert) (News - Alert) 62443, crowned silos of identity, and the ultimate sprawl of OT/IT.
The bottom line is to have “secure-by-design,” and abundant, stable, commercially attractive low-carbon power is increasingly shifting to the “must have” category from “nice to have” status. To stay competitive, we need two things at hyperscaler standards: always-on clean energy and built-in security.
CF: Besides energy, there is the use of water. Are there AI/IoT products you want to point out or missing items that we need to develop to manage water and water treatment plants?
CA: In the sustainability strategy, water has been classified as a formal KPI. Companies globally track water intake and discharge, managing projects within the holding of the Global Water Council. This is a KPI, but the primary focus is on energy efficiency. Some of the AI/IoT products for water saving are integrating flow, pressure, and quality sensors (including pH, turbidity, and conductivity) with edge analytics for real-time anomaly detection. AI can predict leaks, detect abnormal consumption, and control cooling towers to minimize make-up water and blowdown.
From 2024 to 2032, the AI-enabled water management market is projected to grow from $7.5 billion to $53.8 billion, driven by the increasing adoption of predictive analytics, IoT, and water resources management frameworks. Machine learning can dynamically control chemical dosing in treatment plants in real-time, based on sensor readings, thereby minimizing chemical waste and ensuring compliance with regulations. Artificial intelligence tools model hydraulic networks (digital twins) to optimize sensor placement and catch leaks early, reducing non-revenue water and preventable outages. AIoT can make water networks as smart as our furnaces, detecting leaks, forecasting treatment requirements, and closing the loop on reuse.
CF: What other sustainability issues do we face that can be improved with the use of AI?
CA: Absolutely, aside from energy and water, there are several high-impact sustainability levers where IoT and AI can drive change quickly for us. Reducing scrap generation and increasing the recycling of materials through the circular economy are other sustainability challenges that can be addressed using AI. Ultimately, these levers also tie back to energy efficiency.
CF: Do you see particular vertical markets that are past the early adopter phase and taking AI to the next step of innovation?
CA: Good question. Several industries have progressed beyond the proof-of-concept stage and are now viewing AI as a strategic advisor. In the case of industrial manufacturing, a few of the most significant scaled use cases include predictive maintenance, quality inspection, and energy use optimization for value capture.
In healthcare, firms employ AI with IoT sensors for monitoring diabetes patients with AI insulin pumps, as well as for tracking and managing patients’ health status. In retail, organizations monitor real-time inventory levels using computer vision and IoT sensors and implement AI with IoT to enable cashier-less payment systems. In transportation and logistics, firms implement IoT sensors and AI to optimize and calculate routes, which saves a significant amount of fuel. In agriculture, firms implement IoT sensors for precision farming and yield optimization.
CF: What cybersecurity issues do you feel we need to continue to innovate or make a strong commitment to deploy?
CA: The rise of ransomware in Industrial Control Systems (ICS) environments is concerning; 42% of OT incidents result in operational outages, which pose a significant safety concern.
In my view, especially as we are merging OT and IT, the need for a fully converged Zero Trust system is critical. With that, the increased attack surfaces from remote access are concerning. For all users, vendors, and service accounts, mandate identity-based access, including MFA (News - Alert) and conditional access policies. Additionally, apply Zero Trust to cloud and edge workloads, particularly for AI/IoT deployments.
Cybersecurity is more than perimeter-based defenses; it is about identity, integrity, and the intentions of users. What we need is Zero Trust everywhere, AI-governed access security, and commensurate OT visibility to IT rigor in scaling AI and IoT.
Don’t miss Chirag Agrawal’s keynote at AIoT World Expo, taking place February 10-12, 2026 at the Broward County/Greater Fort Lauderdale Convention Center in Fort Lauderdale, Florida. His keynote will take place Wednesday, February 11, 2026, 11:30am. AIoT World Expo is the premier event for exploring the convergence of Artificial Intelligence (AI) and the Internet of Things (IoT) and serves as the gathering point for industry professionals to discover advancements, market opportunities, and understand the transformative power of AIoT across industries. Register today for AIoT World Expo, part of the #TECHSUPERSHOW.