At Adobe’s 24th annual Max conference in Miami this week, the atmosphere was markedly different from the frantic, tech-heavy presentations that have defined the company’s recent trajectory. As I sat in the keynote, observing the demonstrations and listening to the executive team, it became impossible to ignore the delicate tightrope Adobe is currently walking. The company is grappling with a challenge that mirrors the broader struggle within the media industry: the tension between the push for aggressive AI adoption and the reality of a growing, vocal base of skeptical users who are increasingly pushing back against the automation of creativity.

In the tech world, the pressure to "skate to where the puck is going"—a classic Wayne Gretzky-inspired mantra—has driven Adobe to position itself as an AI-first entity. Yet, that ambition is currently colliding with a hostile, or at the very least, wary, crowd. This conflict defines the modern software landscape: companies must innovate to stay relevant, but they must also contend with a customer base that is increasingly protective of human artistry and wary of the ethical and practical implications of generative AI.

The Shift in Tone at Max 2025

For the past two years, Adobe Max has been dominated by a singular narrative: the total transformation of creative work through artificial intelligence. In 2023, the focus was squarely on Firefly, the company’s proprietary image generation model, and the integration of generative capabilities directly into flagship tools like Photoshop and Premiere Pro. By 2024, the rhetoric reached a crescendo; the company famously declared to the assembled audience that "2024 will go down as the year that creativity changed forever."

The 2025 iteration of the conference, however, told a more nuanced story. While AI remained present, the fervor had shifted toward a model of "choice." Rather than a relentless march toward a fully automated future, the keynote presentations featured a lineup of third-party models from industry giants like Google, OpenAI, and Runway. Adobe positioned itself not just as a developer of AI, but as a platform—a hub where creatives can select the model that best fits their specific workflow. The only acknowledgment of the skeptics who have grown tired of the AI hype cycle was a passing, almost understated remark: "Of course, you can ignore it all as well."

This year, the tone was distinctly more grounded. In one of the most significant moments of the opening keynote, incoming CEO Anil Chakravarthy made a point of addressing those who have felt alienated by the rapid integration of AI into their professional toolkits. He explicitly stated that users were not obligated to use AI features if they did not align with their creative process. "Some of you don’t want to have anything to do with AI," Chakravarthy said, promising to prioritize "the precision and control that you’re looking for, and the option to not use AI."

This commitment to the "traditionalist" user was further reinforced by the unveiling of several substantial updates that were entirely devoid of AI integration. The debut of a rebuilt color mode in Premiere Pro and the introduction of Dynamic Text in Illustrator served as powerful signals that Adobe still values the core, manual-control features that made its software the industry standard in the first place. While Firefly continues to be a major pillar of Adobe’s strategy, its role has evolved. It is moving away from being marketed as a standalone "magic" engine and is instead being repositioned as a comprehensive creative studio, functioning more like a collaborative workspace—reminiscent of platforms like Figma—than a raw, black-box AI model.

AI-Optional Does Not Mean AI-Free

Despite the overtures to skeptics and the focus on non-AI improvements, it would be a mistake to assume that Adobe is retreating from its long-term vision. On the contrary, the company is doubling down on AI, albeit with a more sophisticated approach that prioritizes integration and agent-based workflows over simple, novelty-driven features.

Why ‘you don’t have to use AI’ is the new AI pitch

During the same keynote, Chakravarthy announced that Adobe would be providing a mechanism for AI services to interact with Creative Cloud applications directly through the model context protocol (MCP) standard. This development is significant because it shifts the paradigm from "AI inside the app" to "AI driving the app." Under this model, a user could leverage the reasoning capabilities of models like Claude, ChatGPT, or Gemini to act as an agent—effectively delegating the mechanical aspects of editing, content creation, and file exportation to an AI that operates directly on the user’s computer. These integrations are slated for private beta testing in early 2027, signaling a multi-year commitment to this infrastructure.

While Adobe has previously offered AI connectors, the adoption of the MCP standard aims to elevate the precision and control available to creators. By allowing these models to "drive" applications like Photoshop and Premiere, Adobe is attempting to solve the primary complaint of professional users: that current AI tools are often too blunt or imprecise for professional-grade work. By building out its own "Creative Agent"—essentially a chatbot interface that understands user intent and executes complex tasks across Premiere, Illustrator, InDesign, After Effects, and Frame.io—Adobe is trying to bridge the gap between human creative direction and machine execution.

The Great Bifurcation of the Creative Industry

The developments at Max 2025 highlight a profound bifurcation that is currently reshaping the creative industries. On one side of this divide are the AI skeptics—creators, editors, and designers who are deeply suspicious of AI-generated content, concerned about copyright issues, and wary of the potential for AI to diminish the value of human skill. These users prioritize the "precision and control" that Adobe’s CEO promised to protect. They see software as a craft tool, not a creative partner.

On the other side are the AI enthusiasts—a growing cohort that views generative tools as an essential evolution of the creative process. For these users, AI is a force multiplier, a way to clear the bottleneck of mundane, repetitive tasks and focus on higher-level conceptual work. They are already actively integrating tools from across the ecosystem, and for them, Adobe’s pivot toward open model support is a welcome development.

Adobe’s challenge, and indeed its strategic imperative, is to keep both groups under one umbrella. By positioning its Creative Cloud suite as a "big tent" that accommodates both the traditional manual workflow and the emerging agent-based, AI-driven future, the company is attempting to mitigate the churn that might otherwise occur if it pushed too hard in one direction.

The media industry, which relies heavily on Adobe’s tools, is watching this transition with intense interest. Whether it is in journalism, graphic design, or film post-production, the debate over AI is rarely about whether the technology works—it is almost always about how it is used and who controls the output. Adobe’s recognition that it must cater to the user who wants "nothing to do with AI" while simultaneously building the architecture for an "agent era" shows a company that has finally realized that the future of software is not just about raw capability, but about user agency.

As the industry moves toward 2027 and the promised rollout of these deeper agent-based integrations, the success of Adobe’s strategy will likely depend on whether it can maintain that balance. If the company can successfully separate the "AI-optional" creative tools from the automated "agent" workflows, it may just find a way to navigate the hostile waters of the current AI backlash. For now, however, the message from Miami is clear: the era of "AI everywhere, all the time" has given way to a more mature, and perhaps more difficult, era of choice.