Adobe MAX Shows Why Creative Judgment Must Guide AI Production
At its MAX conference in Miami Beach last week, Adobe announced updates spanning Creative Agent, Firefly and GenStudio that extend its effort to connect creative production with brand guidance, campaign planning and performance measurement. MAX has traditionally centered on creative professionals and the tools they use, while the enterprise side of Adobe’s portfolio gets more attention at Adobe Summit, where the focus is marketing and customer experience. AI is drawing those two conversations closer together as Adobe links creative work with the brand knowledge, campaign data and production systems surrounding it.
For years, the technology industry’s conversation about content has centered on velocity and scale, much as productivity dominates the discussion of AI adoption more broadly. But in the creative community, speed and volume are not always the goal, even when the brand paying the bills asks for them. Speed matters, but it should not be the only measure of success. My assessment is that craft has to set the priorities for speed and scale. By “craft,” I mean the taste and judgment that decide what is worth making, which is where creators are drawing their own lines around AI. Adobe’s opportunity rests on how effectively its tools support that judgment in making distinctive work. Creators need control over where AI enters their process and what it can change. Enterprise buyers need to evaluate how brand context and campaign evidence inform those decisions, then measure whether the resulting work achieves its objectives.
(Note: Adobe is an advisory client of my firm, Moor Insights & Strategy.)
Adobe Connects Creative Tools With Enterprise Context
I have been following Adobe’s content supply chain strategy across both events for several years. Adobe’s latest announcements bring more AI assistance and brand context into creative applications. The company announced an expansion of Creative Agent across Creative Cloud and Firefly. The agent lets creators describe a goal in plain language while it coordinates Adobe’s tools and AI models. Adobe also previewed local MCP integrations for Photoshop, Illustrator, Premiere and After Effects, slated for private beta in early 2027, that would let third-party AI agents work with native project files.
For enterprises, Adobe described how CX Enterprise Coworker would coordinate AI agents and content workflows through a conversational interface within GenStudio, its content supply chain solution. Adobe also plans further integration of Brand Intelligence — which it describes as learning from written brand guidelines and undocumented institutional knowledge — into Photoshop, Illustrator, InDesign and Firefly Creative Production for Enterprise. The announcement did not specify availability dates for Coworker in GenStudio or those integrations.
Adobe’s Creators Set Different Boundaries For AI
The creator discussions at MAX showed why Adobe needs to accommodate different levels of AI involvement within workflows. During one creator panel session, photographer Marina Williams described using Firefly Boards to develop visual references grounded in her own portfolio. YouTube filmmaker Tejas Hullur uses AI to organize business thinking and locate material in transcripts, while keeping creative writing separate.
Hullur sees other creators pushing to upload more and more. By contrast, he plans only six releases for 2027, giving each production a longer development cycle. That approach puts craft first: The creative ambition determines the schedule, and the creator is free to determine where AI can help with faster execution.
Many creators have mixed feelings about at least some uses of AI, and Adobe acknowledged the uncertainty accompanying the changes to its products. “When something we love changes so quickly, it is unsettling,” said Deepa Subramaniam, vice president of product marketing for creative professionals at Adobe, during the keynote. Adobe stated it would continue developing tools for creators who want to work without AI.
The company’s enterprise tools also enable selective use of AI-driven features at the task level. A team may welcome automated digital cleanup of an audio recording, for example, while wanting to retain direct control over the narrative and final edit.
L’Oréal Carries Creative Direction Into Production
Adobe Creative Park, where attendees can try the latest Creative Cloud features, among other things, is one of my favorite parts of MAX. This year, the L’Oréal experience gave me the clearest hands-on look at how creative control could extend into a repeatable production process. The L’Oréal team took my headshot and put me in the marketer’s role to create an ad (featuring me as the model) within pre-established artistic boundaries.
Before generating the ad, we used CX Enterprise Coworker in a demonstration to review an earlier campaign. The workflow carried forward those findings, including one that said portraits with a natural face and a clear product story had performed best, as well as saved production steps in Firefly Graph and brand guidelines. My appearance remained unchanged, with AI limited to lighting and assembly adjustments.
What impressed me was how the creative direction carried into the execution. A regional marketer adapting an approved campaign offers a practical use case. The central creative team could establish the visual treatment and permitted changes, giving the marketer a defined way to produce a local version of the larger campaign.
Adobe also offers Simulate, a Brand Intelligence capability that uses synthetic audiences to explore potential reactions to creative work before launch. Adobe describes the feedback as directional, but as I played around with it, I could see the potential for using it to inform how audiences might receive the content. That said, buyers should test whether that feedback helps teams identify unclear messages or select executions for live testing. Campaign measurement is still required to establish actual performance.
GenStudio’s Agency Value Depends On Shared Learning
The L’Oréal experience illustrated the potential value of bringing prior campaign learning into production. Extending that approach across agency relationships raises the harder question of which knowledge should be carried forward and what should remain within each organization.
In a one-on-one meeting at MAX, Ravi Duddukuru, who leads product marketing for GenStudio at Adobe, described for me the concerns he sees on both sides: Brands may restrict internal information, while agencies protect their unfinished work. Yet agencies need brand knowledge to make informed recommendations, and brands want campaign learnings to persist when teams change.
Duddukuru identified content review as one avenue for building shared knowledge. Annotations and feedback can preserve some of the reasoning behind creative decisions without exposing the full project environment. That reasoning also needs context. Feedback on one campaign may reflect a particular audience or objective, rather than a permanent brand requirement. Buyers need agreement on which feedback becomes shared guidance, who can revise it and what remains available when a business relationship ends.
Duddukuru described aspects of shared Brand Intelligence as work still in progress. Adobe’s opportunity is to help campaign learning accumulate while preserving creative teams’ authority to decide when a new assignment calls for a different approach. While I believe this is a good thing, I also think that the company should publicly share results from trials that include creative teams and that measure usable output, revision cycles and campaign outcomes.
Why Creative Judgment Must Set Priorities For Adobe’s AI Production Tools
Adobe’s MAX announcements build on a useful division of responsibility: Creative teams establish the direction, while marketers have more scope to adapt and deploy the work. What remains unproven is whether campaign learning can carry across teams and agency relationships without narrowing the creative choices it is meant to inform.
For me, the significance is the broader role that division of responsibility gives to creative judgment as production becomes more automated. Decisions about what is worth making — and which elements must remain intact — become more consequential when they can guide many executions. That is where craft should set the priorities for speed and scale.