Adobe’s AI Agents Move From Editing Help to Creative Automation

Adobe used its MAX conference in Miami to show how AI is moving beyond one-click editing help. The company announced new agents, editing tools, and connections that could let outside chatbots work inside Creative Cloud projects.

The plan covers Photoshop, Premiere, Illustrator, InDesign, After Effects, Lightroom, Express, Frame.io, and Firefly. Adobe is also giving users more control over how its AI works, including the ability to review plans, choose models, set brand rules, and keep project files editable.

Creative Cloud opens its doors to outside AI agents

Adobe is testing local Model Context Protocol integrations for Photoshop, Illustrator, Premiere, and After Effects. The connections would let external agents from OpenAI, Google, and Anthropic use Adobe tools on a local computer, including making direct changes to native project files.

That means a chatbot such as ChatGPT, Claude, or Gemini could work with Adobe projects instead of stopping at a text response. Layers, vectors, bins, and sequences would remain editable, so users could take over from an agent and refine the work by hand.

Adobe described workflows that connect several tools at once. An agent could pull reference documents from Google Docs, build a project in Adobe software, and share the finished work with colleagues or clients in Slack. MCP-capable tools could join the same workflow.

These local Creative Cloud connections are not available yet. Adobe plans to launch a private beta in early 2027, while Deepa Subramaniam, an Adobe vice president, said the company is still deciding which capabilities to expose and will use feedback from users.

“We are still figuring out where and when and how to draw the line,” Subramaniam said. “What we really want to do is to support that creative workflow, how you choose to do it.”

Firefly assistants take on longer jobs

Adobe is also expanding Firefly AI Assistant with tools for tasks that involve several steps. A new Plan Mode lets users review, edit, and approve an agent’s proposed workflow before it begins. The Model Selector lets users choose a model for a project or leave the decision to automatic selection.

Users can set preferred formats, colors, brand guidelines, characters, and products, then apply those choices across projects. Firefly AI Assistant is also gaining a Firefly Boards integration, and the beta is powered by Adobe’s creative agent.

Adobe previewed “always-on” agents for recurring work. Users could assign an agent to organize incoming assets, run quality checks, or prepare content while they focus on another job or step away from the computer. The user can review the agent’s plan and approve the final result.

One example showed an agent turning incoming images into promotional graphics for different social media channels. Adobe has not said when this preview will launch, but it will come to Firefly AI Assistant first.

Another demonstration involved video footage arriving in a project. Firefly AI Assistant offered to organize raw files, apply LUTs, clean up audio, and back up the files after the user approved the plan. The user then asked the assistant to prepare a social media promotion campaign.

New tools target unwanted objects and editing work

Premiere is getting a Generative Remove tool in beta. Users can mask and track an unwanted object, click Remove, and let Premiere fill the area with realistic detail while leaving the surrounding footage untouched. The tool is aimed at distractions such as boom mics and background clutter.

Premiere is also adding sound tools that separate dialogue, music, ambience, and sound effects into individual tracks and elements. Its Color Mode color-editing experience, which launched in beta in April, is part of the new professional color workflow.

The creative agent is expanding inside Premiere, too. It can track masked subjects through a frame, apply effects, and adjust color and exposure across entire clip sequences. Adobe is bringing the agent to the Premiere mobile editing app, where users can make conversational video edits while away from a computer.

Photoshop is receiving an updated Remove Tool that can better detect and rebuild nearby reflections, shadows, and footprints. Quick Compare adds a split-screen slider for reviewing before-and-after changes without jumping between layers. Relight adds movable light sources around the canvas, with controls for hue and brightness.

Lightroom is gaining the blemish removal tool that was available on Lightroom mobile. Desktop users can retouch skin in photographs with a single tap. Open Eyes is also coming to Lightroom in public beta, replacing closed eyes with natural-looking open eyes by matching a reference photo from the image library.

Firefly reaches into design, video, and Google tools

Firefly is adding more ways to create and organize media. A new video tool can assemble clips with images, sound effects, and other elements. Quick Actions, which automate jobs such as making video thumbnails and product ads, are now generally available.

Firefly Design is launching new beta experiences for illustration, image editing, animation, 3D, and print. Firefly Graph helps creators build and iterate on custom, repeatable workflows. In private beta, Firefly Builder gives users a way to prototype websites and apps, with similarities to Figma Make.

Adobe already supports third-party models from Google, Runway, ElevenLabs, Black Forest Labs, OpenAI, Kling, and ByteDance. Topaz Labs, one of Adobe’s newer acquisitions, is releasing Wonder 4 for upscaling and denoising, along with ImageHDR for turning standard images into HDR.

Adobe also announced an expanded partnership with Google. Acrobat is receiving deeper connections with Google Drive and Gmail, including an “edit with Acrobat” option in file previews. Adobe’s creative and productivity tools will connect with GeminiApp, Google Cloud’s Gemini Enterprise, GoogleWorkspace, and AI Mode in Google Search.

Google’s models will also come to more Adobe apps, including Acrobat. Adobe says the connection will help people edit PDFs conversationally, understand document information, and create presentations, resumes, and reports faster. Acrobat will be available from the Gemini chat window, with AI Mode in Google Search planned for the future.

Adobe’s Zorana Gee said the company is building for people across the creative spectrum. “Ultimately, we’re a creative company serving creatives,” she said, adding that Adobe wants to help anybody hone their craft while keeping the professionalism, security, and reliability of its products.

Stay connected via Google News

Follow us for the latest AI News