Skip to main content
shubhojit
shubhojit
Level 2
April 30, 2026

How are enterprises operationalizing GenAI from Adobe Summit into real production workflows?

  • April 30, 2026
  • 1 reply
  • 49 views

It's been about a week since Adobe Summit, and I'm still thinking about all the deep conversations and ideas that we shared.

As a Solution Architect at IBM, one thing stood out to me: we are moving from experimenting with AI to using it in businesses on a large scale.

Some important things I've been thinking about are

  • How businesses are changing the content supply chain (planning → creation → activation → insights)
  • Problems with scaling GenAI (Firefly) beyond pilots, especially when it comes to governance, FinOps, and observability
  • Putting together platforms like AEM, Workfront, and GenAI tools into a connected system
  • Moving from separate tools to AI-driven workflows that work together

Want to hear from other people:

What real-world examples do you see that are being used in production?
What problems are you having with scaling GenAI?
How are you connecting creative tools with business systems?

I would love to learn from other people's points of view and experiences.

    1 reply

    tekmera
    tekmera
    Level 2
    May 1, 2026

    Nice question ​@shubhojit 


    What I'm seeing is more usage of Workfront's internal AI specifically. Customers are
    asking for AI integration across the whole GenStudio stack, rolled up and visualized across the
    product line. For example, the ability to run conversational queries against your data layer and pull
    insights across the stack.

    David Kershaw | Workfront & Fusion Architecture | Tekmera | https://tekmera.ai/work-with-us/adobe-practice
      shubhojit
      shubhojitAuthor
      Level 2
      May 1, 2026

      That makes a lot of sense—and honestly aligns with what I’ve been seeing as well. The moment AI starts getting embedded within each product (like Workfront), the next natural expectation from customers is for it to work across the stack as a unified layer rather than in silos.

      The conversational querying piece is especially interesting because it shifts the interaction model—from navigating tools to interrogating outcomes. But that also raises some real challenges around how the data layer is structured: consistency of schemas across AEM, Workfront, and other systems, access controls, and how much context the AI actually has to generate meaningful insights versus surface-level responses.

      Curious how you’re seeing teams approach that foundation—are they investing more in unifying the data/metadata layer first, or trying to layer AI on top of existing fragmentation and iterating from there?

        Terms & ConditionsAccessibility statement

        Loading footer...