Adobe Creative Agents Come to Gemini and Claude
Adobe is pushing its AI design tools far beyond the Creative Cloud interface by embedding them directly into leading chat platforms like Google Gemini and Claude. At the center of this strategy is the new Adobe creative agent, the technology powering the Firefly AI Assistant and the “Adobe for creativity” connector already live in Claude. An Adobe connector for Gemini is scheduled to roll out in the coming weeks, extending Firefly’s imaging, design and video capabilities into Google’s multimodal assistant. For designers, this means access to core Adobe Firefly integration features—such as generative imaging and layout support—without ever opening Photoshop or Illustrator. Instead of treating chatbots and creative software as separate steps, Adobe is effectively merging them into one conversational workspace, turning AI assistants into front doors for its generative design software.
Designing in Natural Language, Inside a Single Conversation
The most significant workflow change is that designers can now describe what they want in natural language and have Adobe’s AI agent execute the steps behind the scenes. In Gemini or Claude, you might outline a creative concept—say, a cinematic portrait series or a set of social banners—and the agent orchestrates Firefly’s AI design tools in sequence. Crucially, it pauses for approval at each stage, allowing you to refine prompts, adjust compositions or tweak styles before moving on. This conversational loop keeps ideation, generation and revision inside one chat thread, replacing the old pattern of copying prompts into separate apps and re-exporting assets. The result is a smoother, more iterative process where feedback becomes part of the prompt itself, and the AI acts more like a collaborative assistant than a one-off image generator.
Multimodal Creation and Cross-Platform Workflows
Because these integrations sit inside multimodal platforms, they unlock text-to-image, design and even video creation across multiple AI environments. Gemini creative features will let users move from a written brief to generated visuals, then into layout and basic video concepts, all via the same interface. In Claude, the existing connector already exposes more than 50 tools, enabling tasks like generating multiple platform-ready crops or formats from a single source image. One creator cited being able to output different versions for various channels in a single workflow, instead of manually recreating each variant. Meanwhile, the Firefly AI Assistant remains the most feature-rich hub, with over 60 professional-grade capabilities across Creative Cloud apps. Designers can fluidly choose where to start—Firefly, Claude or Gemini—while still tapping into the same underlying generative design software and asset pipeline.
From Manual Tasks to Agentic AI for Creative Work
Underneath these integrations is a broader shift toward agentic AI—systems that can autonomously execute multi-step creative tasks once given a clear goal. Adobe’s creative agent is designed to connect multiple tools in sequence, reducing repetitive work such as resizing, reformatting and re-exporting assets for different platforms. A photographer, for example, used Firefly AI Assistant to turn simple portraits into more cinematic images without hopping between separate applications. For designers, this means more time spent on concept, taste and storytelling, and less on production chores. As Adobe extends its tools into Gemini and Claude, AI assistants increasingly resemble flexible production partners rather than just prompt-driven generators. The strategic takeaway for creatives is clear: chat interfaces are evolving into full-fledged creative workspaces, and learning to brief, critique and iterate with these agents will become a core design skill.