Beyond insight: What comes next for industrial AI?

Posted: June 19, 2026

Coming out of AVEVA World 2026, I found myself reflecting less on individual announcements and more on something broader that ran through the entire event.

It’s clear we’ve made real progress in industrial AI.

Across keynotes, sessions, and customer stories, the theme of industrial intelligence was everywhere. Connecting data, scaling analytics, and applying AI across the industrial lifecycle no longer feels experimental. It’s becoming embedded into how industrial organizations operate.

But what stood out to me most wasn’t just how much capability now exists—although this is impressive!

It was how consistently the conversation focused on connecting and embedding AI across the industrial lifecycle.

Why insight alone isn't enough

For a long time, the central question in industrial AI was whether we could extract meaningful insights from complex operational data.

In many organizations, that challenge is now largely solved.

Today, companies can increasingly predict failures, identify production process inefficiencies, and uncover optimization opportunities across assets and sites. The intelligence layer has matured significantly.

And yet, the friction hasn’t disappeared. It has simply moved.

Today, that friction often shows up in the gap between insight and execution.

Turning insight into consistent, repeatable action across teams, systems, and processes is still where things often slow down. Not only because systems are fragmented, but because organizations need confidence that the information, recommendations, and actions are based on trusted context.

AI is becoming part of the industrial lifecycle

Looking across the agenda at AVEVA World 2026, the story was very consistent. Whether the focus was engineering, operations, maintenance, or sustainability, the same pattern kept emerging: AI is no longer a separate layer sitting on top of industrial systems. It’s increasingly becoming part of how those systems are designed, operated, and improved.

In engineering, generative AI is starting to reshape how work gets done by accelerating design activities, reducing iteration cycles, and helping teams keep projects on time and within budget.

In operations, AI is increasingly becoming part of how organizations monitor, optimize, and continuously improve performance. With AI embedded into operational workflows, organizations can combine data sources, get insights quicker, identify subtle deviations earlier, and anticipate issues before they impact production. Teams gain a clearer understanding of how operational decisions affect throughput, quality, energy use, and other trade-offs, helping them make decisions with more confidence.

Across use cases ranging from centerlining to energy optimization, this creates a more consistent way of operating and helps organizations stay closer to optimal conditions while avoiding unintended consequences.

In maintenance and asset performance, the value is just as tangible. AI is helping organizations move further toward predictive maintenance strategies. Instead of relying only on fixed maintenance intervals or reacting after failures occur, teams can anticipate issues earlier, forecast time to failure, and better understand whether operational adjustments can safely extend performance until the next planned outage.

The result is improved reliability, reduced downtime, and more efficient use of maintenance resources.

Taken together, these shifts point to something broader.

Industrial intelligence increasingly involves connecting analytics, workflows, and people across lifecycle disciplines so information can move more effectively into operational execution.

From automation to intelligence—and now toward impact

Many sessions at AVEVA World described this as a progression from automation to intelligence, and now increasingly toward enterprise scale.

That last step is where many organizations are now focusing their attention.

Because once intelligence becomes embedded into industrial operations, the focus becomes less about what we can know and more about what organizations are actually able to do with that knowledge and how quickly they can turn it into action.

Why agentic AI feels like a natural progression

In that sense, agentic AI doesn’t feel like a sudden leap, but more like a natural continuation of where the industry has already been heading.

At its core, agentic AI is about helping organizations respond more naturally and consistently once something important happens.

But industrial environments are complex. Real workflows are rarely handled by a single capability or a single step. Problems move through detection, diagnosis, interpretation, coordination, and action, often across multiple systems and teams.

At AVEVA, the foundation of our industrial intelligence solutions is CONNECT, which brings together trusted and contextualized industrial data. On top of that sits an agentic framework that helps connect data, workflows, models, and governance across the lifecycle so insights can move more smoothly into operational execution.

From there, agentic solutions like digital twin builder, asset monitoring, and operational optimization combine multiple capabilities into something that can deliver real operational outcomes.

Underneath that are the agents themselves, with task-specific capabilities that continuously detect patterns, support decisions, and help teams respond in real time.

That structure helps organizations scale AI capabilities more consistently across the enterprise.

Value is rarely created by a single insight alone. It’s created when insights move through workflows and become part of operational decision-making.

Making digital twins more practical at scale

Digital twins have been part of the industrial conversation for years, but what feels different now is how much more practical and scalable they are becoming.

For a long time, building meaningful digital twins required significant manual effort. Data needed to be mapped, models aligned, and context continuously maintained across systems. The value was clear, but scaling those efforts was often difficult.

What’s changing now is that much of this work can be completed more efficiently and at greater scale.

With agentic solutions, AI can help structure industrial information more efficiently by bringing together data from different systems, aligning it into usable models, and helping organizations operationalize those models much faster.

The knowledge graph, which functions as the layer that preserves context and relationships across assets, systems, and processes over time, becomes especially important. It helps digital twins reflect not just structure, but operational behavior and meaning.

For agentic AI, that connected context plays an important role in helping systems understand relationships, make decisions, and act more effectively. Insights become trustworthy, actionable and scalable across the organization.

From insight to action without removing the human

One thing I think is important to say explicitly is that none of this is about removing people from the process.

Industrial environments are simply too complex, too critical, and too nuanced for that.

What organizations are really moving toward is something more balanced. Systems that reduce effort, connect information faster, and help guide decisions while still keeping humans firmly in the loop.

Operators and engineers remain at the center of it.

They simply spend less time navigating fragmented information and more time acting with clarity and confidence.

Final takeaway: The power of connected intelligence

Looking back at AVEVA World 2026, what stood out wasn't a single breakthrough announcement. It was the growing alignment across technologies, use cases, and customer priorities.

Industrial AI and data platforms connected. AI embedded across the lifecycle. Digital twins becoming more practical. And a clearer path emerging between insight and operational execution.

It’s a gradual shift, but one that is capable of changing how industrial systems operate in practice.

And that's what makes this shift worth paying attention to. As AI becomes more deeply embedded across the industrial lifecycle, the focus increasingly shifts toward connecting intelligence, context, workflows, and domain expertise in ways that support better decisions and more effective execution at scale.