We went into Opticon 2026 this week expecting to hear a lot about AI and, of course, we did. From Virtual Teammates and agentic workflows to AI-powered experimentation, content operations and search, it was impossible to leave Opticon without an appreciation for just how quickly the technology is advancing.
But that wasn’t my biggest takeaway from the week.
What stood out to me most was how much the conversation around the technology has evolved.
We’re moving beyond simply asking what AI can do and are now having conversations about what people should do differently because of it. We’re talking about how work should change, where human judgment matters most, how organizations need to operate differently and, ultimately, how we ensure all of this new technology leads to improved business performance.
In many ways, Opticon reinforced something we’ve been talking about at Aperture Labs all year:
Technology changes what’s possible. Realizing that possibility requires evolving the work around it.
Here are the four topics from Opticon that stood out most to me:
One of my favorite narratives from the day surfaced in quite a few keynotes and breakout sessions: the most valuable output of AI isn’t the content it creates. It’s the capacity it unlocks.
For the last few years, the AI conversation has focused on efficiency. How much faster can a task be completed? How many hours can be saved? How much more content can be produced?
Those are useful measures, but they stop short of the bigger opportunity.
If AI gives someone five hours back, saving five hours isn’t the business outcome. What the organization does with those five hours is.
Do we simply ask that person to produce more of the same work? Or do we create space for the work that uniquely benefits from human judgment, creativity, expertise and connection?
That’s a question Tina Nelson from Optimizely and I discussed recently, and one I believe every leadership team should be asking as they develop their AI strategy.
A related narrative that echoed throughout Opticon is that humanity is becoming a premium layer.
AI can identify patterns, generate options, analyze enormous amounts of information and increasingly execute work on our behalf. But capability and judgment aren’t the same thing. Someone still needs to decide what matters. What is worth pursuing? What feels right for the customer? Where should we take a risk? Where should AI not be used?
The more capable AI becomes, the more valuable those decisions become.
So perhaps the opportunity isn’t simply to find more work for AI to do. It’s to use AI to create more capacity for the work where people create the greatest value.
That shifts the question from “where can AI do the work?” to “where can we elevate the value of human work?”
This was another theme that had me nodding my head enthusiastically during the sessions, many of which weren’t about technology at all.
Instead, they hit on a reality we see continuously in digital modernization work: the technology is often capable of moving much faster than the organization around it.
When clients first come to us, we consistently see their teams losing valuable time chasing down information and approvals or searching for the latest version of an asset. They’re spending time manually moving information between systems or sitting in status meetings. Often, work is simply moving through processes that were designed years earlier for the technology available at the time.
But as technology changes, so must the way we work.
We can give teams incredibly powerful new technology, but if roles, workflows, governance and decision-making remain unchanged, we shouldn’t expect the technology alone to transform performance.
The evolution from AI assistants to AI agents and Optimizely’s “Virtual Teammates” was another major theme at Opticon.
This evolution is important because it changes the role of AI within an organization. We’re moving from AI helping to complete a task to AI performing work across an entire process.
Naturally, this shift introduces an entirely different set of questions. What information can an agent access? What decisions can it make? When does work move from an agent to a person? What happens when multiple systems or agents are involved? Who is accountable for the outcome? Where do approvals live? How do we govern all of it without recreating the very bottlenecks we’re trying to eliminate?
We’ve moved beyond asking integration questions and now must ask orchestration questions.
Connecting systems and moving data between them remains important, but connected technology doesn’t necessarily create connected work. As AI becomes capable of doing more of the work itself, organizations will need to become much more intentional about designing how people, systems, data and AI work together.
Experimentation, while always a big theme with Optimizely, felt particularly relevant at Opticon this year.
We can all agree that AI dramatically reduces the time and effort required to create, analyze and iterate.
When trying something new was expensive and slow, organizations had good reason to spend a significant amount of time attempting to predict the right answer before acting. As experimentation becomes faster and cheaper, the equation changes.
The goal becomes less about perfectly predicting the future and more about building an organization capable of continually learning its way toward better outcomes.
Hypothesize. Act. Measure. Learn. Adapt. Then do it again.
Every major shift in digital technology has required organizations to become comfortable with some degree of uncertainty. AI is no different.
The organizations that realize the greatest benefits from AI won’t necessarily be the ones that develop the perfect AI strategy today. More likely, they’ll be the ones that build the systems, governance and operating models that allow them to keep learning as the technology evolves.
There was plenty of impressive technology shown at Opticon and the pace of advancement is undeniably extraordinary.
But I left thinking less about the technology itself and more about what needs to change around it if we want to capture its full value.
For me, that was the real takeaway from Opticon. The technology is going to keep getting better (and probably faster than any of us can fathom). The bigger challenge for organizations will be making sure the way they work keeps getting better, too.