CEO, Zinnov | Building GCCs for Global 2000 and private equity firms | Rewiring tech services companies around an AI core | Designing and scaling channel partnerships for AI labs and hyperscalers
I have written before about three stages of AI adoption: individual productivity, enterprise productivity and frontier capability, where AI does what was not possible before. Most companies are still celebrating stage one, while physical AI, which puts intelligence into buildings, plants and the products themselves, is already one of the frontiers. I saw this up close at the first Fortune AIQ Summit at the NYSE, where Suresh Venkatarayalu, CTO of Honeywell Technologies, spoke with AJ Wijesinghe of Ecolab and Emily Forlini of Fortune. Suresh made the case clearly. Physical AI moves buildings and plants from following fixed rules to systems that see, think, act and learn, so they can spot problems before equipment fails, help operators make faster decisions and capture decades of engineering knowledge that would otherwise retire with the people who hold it. The bar is high, though: industrial customers expect 99.9999% reliability, while today’s best AI models may be closer to 85%, which is why Honeywell is moving from automation to autonomy step by step, with operators in the loop as trust builds. This week, Honeywell launched its physical AI product, Forge Cognition, built in deep collaboration with NVIDIA and Dell, and went from first conversation to launch in about nine months. It brings together AI agents, live digital twins and decades of Honeywell engineering knowledge so that buildings and plants can understand what is happening, spot problems early and act alongside the people who run them. In the pilot at NVIDIA’s Voyager building it cut unnecessary alerts by up to 90% and found up to 10% potential energy savings, while keeping occupant comfort above 95%. My view is that physical AI will be won by the companies that know the machines and bring the best AI to them, which is exactly what Honeywell is doing. Congratulations to the Honeywell Technologies team. #PhysicalAI #FortuneAIQ #Honeywell Zinnov
Physical AI changes the operating model more than the technology stack. The challenge is not deploying intelligent systems into plants or buildings. It is redesigning governance so humans and autonomous systems make decisions without creating new operational bottlenecks. Long-term advantage will come from execution discipline, not AI capability alone.
Great example Pari and there is so much potential in physical AI. Germany is about to ramp up a lot to get this phase of the development right and there’ll be hopefully many more important developments to share in the future.
Honeywell has deep industrial experience, especially with Automation business
Great perspective !
well said
The 99.9999% reliability bar is the crux of it — industrial customers have zero tolerance for model variance, which is why physical AI will be won by the companies that know the machines, not just the models. Congrats to Suresh and the Honeywell team on the Forge Cognition launch.