Augmented Ops is a podcast for industrial leaders, innovators, and operators shaping the future of frontline operations.
In this episode of The Augmented Ops Podcast, Roey Mechrez, Tulip’s Head of Ecosystem, sits down with Erik Udstuen, CEO and Co-Founder of TwinThread, to unpack one of manufacturing’s most talked-about concepts — digital twins — and explore what it really takes to bring industrial AI to life.
Their conversation comes at a pivotal moment for the industry. AI-driven operations are no longer a distant goal but a present-day differentiator. Across sectors, manufacturers are working to merge physical systems with digital intelligence — and, in the process, redefine what operational excellence looks like.
For years, “digital twin” has been used as shorthand for a digital copy of a machine or process. But as Erik explained, that definition undersells the real opportunity. The most valuable digital twins aren’t static models — they’re living systems that learn, evolve, and respond to real-world data in real time.
In his view, the next generation of digital twins are better described as learning twins: simulations that continuously adapt as conditions change, allowing manufacturers to test ideas virtually before implementing them physically.
This evolution represents a shift from simply representing reality to actively improving it. When combined with AI, digital twins enable teams to forecast performance, optimize energy use, and identify anomalies long before they impact production.
Throughout the discussion, Erik and Roey explored the difference between systems that inform and systems that act.
Predictive analytics has long been part of manufacturing’s digital transformation story — alerting engineers when something might go wrong. But AI and simulation together are creating prescriptive systems that can identify issues, suggest corrective actions, and even implement fixes autonomously.
These capabilities blur the line between automation and intelligence. As Erik pointed out, the future isn’t about gathering more data but about giving that data context — connecting operational variables to real business outcomes.
AI becomes the engine that turns information into insight, bridging the gap between what’s happening and what should happen next.
Even as automation advances, both Roey and Erik emphasized that people remain central to manufacturing’s digital future.
AI doesn’t replace human expertise — it amplifies it. Digital twins and AI models depend on engineers, operators, and domain experts to define parameters, validate outputs, and interpret results.
“AI doesn’t remove the human from the loop,” Erik said. “It raises the value of human judgment.”
The real opportunity lies in building systems that make human expertise more scalable. The best use cases for AI and digital twins don’t eliminate human decision-making; they make experience transferable and repeatable across sites, teams, and generations of workers.
As the conversation wrapped, Erik described the next phase of industrial AI as one of connectivity and learning.
Today’s most forward-looking manufacturers aren’t just collecting more data — they’re connecting it. When systems, models, and people share context in real time, AI can continuously refine itself, learning from every production cycle and feedback loop.
This convergence of digital twins and industrial AI is what enables continuous improvement at scale. Each cycle of learning makes operations more resilient and efficient, while keeping people at the center of decision-making.
Industrial AI and digital twins are converging toward the same goal: faster learning, safer decisions, and smarter operations. But as Erik and Roey made clear, the future of manufacturing isn’t just about automation — it’s about augmentation.
The factories of the future won’t just react faster; they’ll think faster. And behind every model and insight, there will still be people — engineers, operators, and innovators — shaping how intelligence gets applied.
🎧 Listen to the full episode and explore more conversations with industry leaders at AugmentedOps.com.
