Adobe Adds AI Agents and MCP to Photoshop, Premiere, and Firefly

Adobe (ADBE) unveiled new AI agents and a local Model Context Protocol at its MAX event in Miami on Tuesday, pouring more AI into Photoshop, Premiere, Illustrator, and InDesign and expanding its Firefly creative tool.

The company is making its agent proactive and always on so it can gather more context about work and be present for any action, and it is adding quick actions such as removing a background or generating a thumbnail for a video. Firefly is gaining a private-beta prototyping tool called Builder, a tool for creating on-brand graphs, and an image-editing tool that can change part of an image without affecting other objects. A new AI-powered video tool in Firefly will let users assemble a video from clips, images, and sound effects.

Photoshop is getting a new relighting feature and sliders for prompt-based generative styles; Premiere will gain a color grading system for professionals and an object removal feature for video; Lightroom will be able to correct blinking eyes and remove blemishes. Adobe already offers third-party models from Google, Runway, ElevenLabs, Black Forest Labs, OpenAI, Kling, and ByteDance, and its Topaz Labs unit is releasing two image models, Wonder 4 for upscaling and denoising and ImageHDR for turning standard images into HDR.

Adobe already had ChatGPT and Claude integrations for some product functions; the new local MCP exposes more features of its Creative Cloud apps to users. Adobe vice president Deepa Subramaniam told TechCrunch the company is still figuring out which capabilities to expose and will rely on user feedback.

Adobe shares were down 0.26% in delayed trading.

The company has already added AI features across Photoshop, Express, Premiere, and Illustrator, and at MAX it continued that push. Last month, Figma co-founder Dylan Field said at an event in India that MCP is not a perfect standard and benefits frontier labs more.