Adobe adds more AI to its products and launches an MCP

Summary

• Adobe announced major AI feature updates for Photoshop, Premiere, Illustrator, & InDesign at its MAX event in Miami.
• The company is expanding Firefly's creative AI agent capabilities & releasing a Model Context Protocol (MCP) to integrate Adobe features into other chatbots like ChatGPT, Claude, & Gemini.
• Firefly is entering Figma's market with a private beta prototyping tool called Builder, a new on-brand graph creator, & an image editing tool that modifies specific parts of an image without affecting other objects.
• Firefly is launching an AI-powered video tool to assemble videos from clips, images, & sound effects, while making its agent proactive & always on to provide context for any action.
• Quick action features include background removal & video thumbnail generation.
• Photoshop receives new relighting features & sliders for prompt-based generative styles.
• Premiere gains a professional color grading system & an object removal feature for video.
• Lightroom adds tools to correct blinking eyes & remove blemishes.
• Adobe integrates third-party models from Google, Runway, ElevenLabs, Black Forest Labs, OpenAI, Kling, & ByteDance.
• Acquired company Topaz Labs is releasing two image models: Wonder 4 for upscaling & denoising, & ImageHDR to convert standard images to HDR.
• Adobe is releasing a local MCP to expose more Creative Cloud app features to users.
• Adobe vice president Deepa Subramaniam stated the company is determining which capabilities to expose based on user feedback & aims to support creative workflows without limiting MCP endpoints.
• Figma co-founder Dylan Field previously criticized MCP as an imperfect standard that benefits frontier labs more.