Adobe Releases AI Agents

Adobe today released AI agents to help businesses build, deliver, and optimize customer experiences and marketing campaigns.

Powered by the Adobe Experience Platform (AEP) Agent Orchestrator, Adobe is also creating an AI platform for businesses to manage and customize agents from Adobe and across third-party ecosystems.

"Adobe has long helped businesses deliver engaging experiences to their customers, by turning digital data into actionable insights. We are now leveraging agentic AI to build specialized agents and embedding them into data, content and experience creation workflows," said Anjul Bhambhri, senior vice president of engineering for Adobe Experience Cloud, in a statement. "Our agentic AI innovations are elevating customer experience orchestration by reimagining processes, unlocking productivity for marketing teams and delivering personalized experiences at scale to drive growth."

Adobe also released AEP Agent Orchestrator, including a reasoning engine where decision science and language models will enable dynamic and adaptive reasoning. This will ensure user intent can be interpreted from natural language prompts, to dynamically determine which agents are activated as part of an orchestrated plan. With AEP as the foundation, Agent Orchestrator will drive contextually relevant and goal-oriented automated actions, with support for refinement using a ;human-in-the-loop approach.

Out-of-the-box AI agents, which will be surfaced directly within Adobe enterprise applications such as Adobe Real-Time Customer Data Platform, Adobe Experience Manager, Adobe Journey Optimizer and Adobe Customer Journey Analytics, can be leveraged by businesses to enhance the skills of marketers and accelerate Customer Experience Orchestration (CXO). These AI agents include the following:

• Audience Agent, to create, scale, and optimize audiences that can be activated for personalization initiatives. High-value audiences can be leveraged through actionable recommendations and monitored closely to meet organizational goals and KPIs.
• Journey Agent in Journey Optimizer for the creation and orchestration of customer journeys and campaigns across channels such as web, mobile, app, email, and more. The agent will create journeys based on defined goals, optimize touchpoints based on aspects such as customer drop-off and uncover insights to fine-tune interactions.
• Experimentation Agent, which analyzes experimentation performance data, enabling teams to hypothesize new ideas for optimization, analyze causal impact, and identify predicted conversion or lift. It will be available in Journey Optimizer Experimentation Accelerator.
• Data Insights Agent in Customer Journey Analytics (CJA), ehich streamlines the process of deriving insights from signals to visualize, forecast, and remediate customer experience initiatives.
• Site Optimization Agent, which delivers always-on support for teams to manage brand websites for high performance. The agent automatically detects and raises issues impacting customer engagement, such as broken back links or low-performing pages that can then be addressed via Adobe Sites Optimizer.
• Product Support Agent, which enhances the process of resolving issues, leveraging knowledge sources and organizational data. This now includes troubleshooting support, including case creation and tracking, directly in user workflows.

Coming soon, Experience Platform Agent Composer provides a single interface for businesses to customize and configure AI agents based on brand guidelines, organizational policy controls, and more. This will enable teams to fine-tune AI agent actions and shorten time to value. Additionally, new developer tools, including an Agent SDK and Agent Registry, will enable developers to build, extend, and orchestrate agentic apps, expanding use cases into new industries and user personas. Agent Composer equips businesses with tools to drive multi-agent collaboration using the Agent2Agent protocol.

Adobe also announced new agentic AI partnerships with Cognizant, Google Cloud, Havas, Medallia, Omnicom, PwC, and VML, enabling seamless execution of workflows across agents, as well customization across industries and use cases.