Adobe Unveils New AI Tools and Agents Across Creative Cloud Apps at MAX Event

Adobe has announced a series of updates integrating proactive artificial intelligence agents and new creative tools across its core software lineup, including Photoshop, Premiere, Illustrator, and Firefly [1]. The announcements were made during the company's MAX event held in Miami [1].

Expansion of Firefly and Prototyping Tools

Adobe is introducing a new prototype builder tool within Firefly currently in private beta, which enters into competition with Figma [1]. The Builder tool enables users to create designs for showcasing concepts [1]. Additional additions to Firefly include a tool for generating on-brand graphs and an image editing feature that permits users to alter specific parts of an image without changing other objects [1]. Furthermore, a new video tool powered by AI will allow users to assemble videos using clips, images, and sound effects [1].

Application-Specific Updates

Across its broader software ecosystem, Adobe is rolling out multiple workflow enhancements. Photoshop is receiving a new relighting feature and prompt-based generative style sliders, while Premiere introduces a professional color grading system and video object removal [1]. Lightroom is also gaining blemish removal and an option to correct blinking eyes [1].

Adobe is introducing an always-on, proactive AI agent designed to gather context regarding ongoing work and remain available for various tasks [1]. The company is also adding capabilities from recent acquisition Topaz Labs, which is releasing two image models named Wonder 4 for denoising and upscaling, and ImageHDR for converting standard images into HDR format [1].

Model Context Protocol and Interoperability

Adobe is releasing a local Model Context Protocol (MCP) to allow third-party chatbots, such as ChatGPT, Claude, and Gemini, to utilize Adobe features [1]. This local MCP exposes various endpoints and capabilities across Illustrator, Photoshop, After Effects, and Premiere [1]. Adobe Vice President Deepa Subramaniam stated that the company is currently evaluating user feedback to determine which features and capabilities to expose [1].

"We are still figuring out where and when and how to draw the line. What we’re putting out there is a number of endpoints and capabilities across Illustrator, Photoshop, After Effects, and Premiere. We’re just going to see how people use it, what they’re trying to do, and keep showing more. We don’t have any need or desire to sort of limit what the MCP endpoint can expose. What we really want to do is to support that creative workflow, how you choose to do it," Subramaniam said [1].

In contrast, Figma co-founder Dylan Field commented at an event in India last month that MCP is imperfect as a standard and tends to favor frontier AI labs [1].

Adobe offers various third-party models from companies like Google, OpenAI, Runway, ElevenLabs, Black Forest Labs, Kling, and ByteDance [1]. Prior to the MAX event, integrations with ChatGPT and Claude already allowed users to access select product functions [1].

The introduction of the Model Context Protocol and proactive AI agents by Adobe represents a significant shift in how third-party conversational assistants interact with professional creative software, potentially streamlining workflows across the global digital media industry.

The exposure of Creative Cloud features via MCP could lead to deeper integration between professional design workflows and consumer-facing AI models, changing how digital creators execute repetitive or multi-app tasks.

Adobe is still determining the full extent of capabilities it will expose through the MCP endpoints based on user feedback, leaving the final scope of integrations open.