Every year, marketers gather at Adobe Summit for an insider’s view into the future of digital experiences—and the new tools to help marketers create those experiences.
This year, the work of Adobe Research was behind some of Summit’s biggest announcements, including new AI agents built into the Adobe Experience Platform (AEP), and sneak previews of tools that will help marketers design entire campaigns, swap images seamlessly into catalog spreads, combine real-world info with their own data, and more.
Here’s a look at a few of the big contributions from Adobe Research at this year’s Summit.
Agentic AI is coming to Adobe Experience Platform
Last year, the new AI assistant for AEP was one of the most exciting Summit reveals. Developed with technology from Adobe Research, the AI assistant allows marketers to use natural language to ask questions about their data. As it turns out, that was just the beginning.
This year, Senior Vice President for Experience Cloud Engineering and keynote speaker Anjul Bhambhri unveiled the next big AI advancement inside AEP — a new collection of AI agents, each designed with specific capabilities that will help marketers deliver personalized experiences at scale.
Four of the ten purpose-built, out-of-the-box agents currently in development are backed by major contributions from Adobe Research:
- The Audience Agent analyzes engagement data to create goal-based audience segments for personalization campaigns.
- The Data Insights Agent gathers insights from across an organization so marketers can visualize, summarize, and enhance customer experiences.
- The Experimentation Agent allows marketers to brainstorm new experiments, seamlessly run them, and draw meaningful conclusions.
- The Product Advisor Agent supports brand engagement through product discovery experiences based on an individual’s preferences and past purchases.
As Bhambhri explained on stage, AEP users will engage with the new AI agents conversationally, just as they do with the AI assistant. “But they [the AI agents] go beyond answering questions,” she explained. “They can actually do things for you… You, as the practitioner, are specifying what you want and the agent focuses on the how.”
Users will engage with the AI agents by way of another innovation, the Adobe Experience Platform Agent Orchestrator, which is being built with key contributions from Adobe Research. The Agent Orchestrator has a reasoning engine that allows AI agents to perform decision-making and problem-solving tasks, and it’s where marketers can build, manage, and orchestrate their Adobe AI agents, as well as agents from third-party ecosystems.
Adobe also unveiled Brand Concierge —an AI-powered product discovery solution that enables enterprise brands to deliver personalized, conversational experiences to their customers. Built on Adobe’s Experience Cloud and grounded in deep consumer and customer research, Brand Concierge helps users explore, compare, and select the right products or services through natural language interactions. It supports both text and voice input, and generates brand-aligned, data-driven responses that can include personalized product details, offers, and visual content. The Adobe Research team helped develop the underlying agentic AI system that powers this experience.
“Adobe Research has been working behind the scenes on AI assistants for a long time, so it’s exciting to see agentic AI become a part of the Adobe Experience Platform—and to see marketers’ enthusiastic responses at Summit,” says Vishy Swaminathan, Senior Principal Scientist and the leader of the Research organization behind Agentic AI within Adobe Research. “These new agents can reason and collaborate with each other to help solve Adobe customers’ problems.”
Keynote speaker Shiv Vaithynathan, Vice President, Platform Engineering and Architecture, helped demonstrate the deep reasoning capabilities of AI agents with an onstage demo, taking audiences through the process of creating a tailored audience for an email campaign to introduce a new soda flavor. Audiences got to see how a human marketer will interact with the new Audience Agent, and how the Audience Agent explains the deep reasoning behind every decision.
Sneaks gave a peek at cutting-edge technology in development
Sneaks are a chance for the Summit audience to glimpse exciting experimental technology that could become part of Adobe’s products—and they’re a highlight every year. This year celebrity host Ken Jeong and Adobe host Eric Matisoff kicked off Sneaks by trying out a new tool, developed by Adobe Research and powered by Adobe Firefly, that turns a user’s hand-drawn sketch into a detailed image.
Four of the main-stage Sneaks were also built with innovations from Adobe Research:
Project Slide Wow, presented on stage by Adobe Research Scientist Jane Hoffswell, turns the data in Customer Journey Analytics (CJA) users’ dashboard into stories, and instantly transforms those stories into Microsoft PowerPoint presentations, complete with visualizations and speaker notes. Users can even ask Project Slide Wow questions and make instant updates using the latest numbers, saving time and effort.
Adobe Research and the CJA team collaborated to create Project Perfect Context, which allows users to choose from a wide range of real-world insights. Then, an AI agent connects those insights—such as world news, market trends, and weather—to a team’s own data to help make sense of customer behavior.
Adobe Researchers teamed up with the AEM (Adobe Experience Manager) team to create Project Get Savvy, a tool that lets marketers choose a team of AI agents to help them every step of the way through a campaign, from ideation through developing a full marketing strategy and creating compelling content and visuals.
Adobe Researchers, along with the Adobe InDesign team, created Project Vision Cast, which lets users combine data insights and explore brand imagery, fostering creative brainstorming backed by the numbers. A key element of Project Vision Cast, developed by Adobe Research, is innovative generative AI that can place a product image into any catalog spread, automatically adjusting its angle, lighting, and shadows.
Sessions, labs, and a mobile app that tied the whole event together
The Summit labs and sessions helped attendees learn more about how to use Adobe’s tools, and several focused on Adobe Research’s work. At the Get Inspired and Scale Faster with Generative Experiments session, Adobe Research Scientist David Arbour and Principal Product Manager Paul Aleman helped attendees learn how to use advanced AI-powered technology to experiment and identify ways to increase the impact of digital experiences. The sessions on AI-Driven Analytics: Adobe’s Vision for Customer Journey Analytics and Data Collaboration with Adobe also featured technology from the Adobe Research team.
Adobe Researchers also collaborated with the AEP AI Assistant team to develop Summit Concierge, the mobile app that helped guide attendees through the entire event with scheduling, announcements, maps, and more.
You can catch up on everything from this year’s Adobe Summit here.
Wondering what else is happening inside Adobe Research? Check out our latest news here.
