Adobe debuts CX Enterprise agentic AI platform

At Adobe Summit 2026 in Las Vegas, Adobe unveiled CX Enterprise, an end-to-end agentic AI system designed to manage the full customer lifecycle, from acquisition through loyalty. The company positioned the platform as a way for brands to move beyond isolated AI experiments toward continuous, orchestrated customer engagement.

CX Enterprise brings together AI agents, reusable agent skills and developer tools to simplify how businesses manage consumer lifecycles. It is grounded in Adobe’s Experience Platform (AEP), which Adobe says now powers more than one trillion experiences per year for over 20,000 global brands.

Shares of Adobe rose 2.2% in morning trading on April 20, but the stock remained about 30% lower year-to-date amid a broader selloff in software names linked to AI-driven automation concerns.

Adobe introduced CX Enterprise Coworker as an agentic AI layer that orchestrates customer experience workflows across AEP and applications such as Real-Time CDP, Customer Journey Analytics, Journey Optimizer, Marketo Engage and Target. The system is fully agentic, monitoring signals, recommending next-best actions and executing experiences across channels in real time with human oversight.

Coworker is designed to translate business goals into multi-step actions. Adobe highlighted scenarios such as a marketing team targeting a specified uplift in cross-sell performance, where Coworker would assemble audience segments, creative assets and performance insights, then help execute and monitor the campaign once approved.

The offering is built on open standards including Model Context Protocol and Agent2Agent, and is expected to be generally available in the coming months. Adobe is also integrating NVIDIA’s (NVDA) OpenShell secure runtime and Nemotron open models to support governed agents, including in regulated industries.

Within CX Enterprise, Adobe announced Adobe Brand Intelligence and Adobe Engagement Intelligence. Brand Intelligence is a continuously learning engine that absorbs feedback, annotations, approvals and rejections to capture nuanced brand identity that AI agents can use for content production. Engagement Intelligence is a decision engine optimized around customer lifetime value to determine next-best offers and actions.

Adobe also expanded its GenStudio content supply chain solution with an “agentic” architecture that connects planning, creation, activation, delivery and insights. New capabilities include workflow optimization agents in Adobe Workfront, an AI-driven campaign brief canvas, Firefly Creative Production for Enterprise Workflow Builder, and 3D digital twin workflows developed with NVIDIA (NVDA).

Additional GenStudio modules include GenStudio for Content Marketing, support for ChatGPT Ads in GenStudio for Performance Marketing, and an agency system of record intended to maintain governance and shared brand context across internal teams and external agencies.

Adobe introduced a brand visibility solution as an expansion of Adobe Experience Manager to address how brands appear across AI-driven discovery surfaces and owned channels. The solution is structured as a continuous “sense, generate, reach and learn” operating model for optimizing brand visibility and engagement.

Key components include Adobe LLM Optimizer and Adobe Commerce enhancements to assess and improve AI-era product and content visibility, as well as new agentic authoring capabilities in AEM. Agents such as Brand Experience Agent, Content Advisor Agent and Brand Governance Agent are designed to update sites, surface approved content and enforce brand policies.

Enhancements to Adobe Brand Concierge and new LLM Apps in AEM are intended to bring conversational experiences and branded applications into large language model interfaces, while feeding human corrections back into the system to strengthen governance and performance over time.

Adobe emphasized an open, interoperable ecosystem around CX Enterprise. AI agents, skills and developer tools are being made available within platforms from Amazon (AMZN) Web Services, Anthropic, Google Cloud, IBM, Microsoft (MSFT) and OpenAI, allowing Adobe’s customer experience intelligence to surface inside tools such as Microsoft (MSFT) 365 Copilot, Amazon (AMZN) Quick, Anthropic Claude Enterprise, ChatGPT Enterprise, Gemini Enterprise and IBM watsonx Orchestrate.

The company deepened its collaboration with NVIDIA (NVDA), which is providing Nemotron models and Agent Toolkit software as well as the OpenShell runtime to run secure, policy-governed agentic workflows on premises or in the cloud. A cloud-native 3D digital twin solution built on NVIDIA Omniverse and OpenUSD is now generally available for scalable product content creation.

Adobe is also expanding ties with agencies including Omnicom, Publicis, WPP, Dentsu, Havas and Stagwell, which are standardizing on CX Enterprise to co-develop solutions. System integrators such as Accenture (ACN), Capgemini, Cognizant, Deloitte Digital, EY, IBM, Infosys (INFY), PwC and TCS are packaging agentic solutions for key industries.

Ahead of Adobe Summit, RBC Capital on April 16 lowered its Adobe price target to $350 from $400 due to multiple compression across peers but maintained an Outperform rating. The firm expects management to stress ecosystem strength and the return on investment focus of the company’s generative AI offerings, while investors seek re-acceleration in annual recurring revenue.

BTIG initiated coverage on April 12 with a Neutral rating and no price target, citing uncertainty around how AI will reshape the creative ecosystem and Adobe’s long-term revenue and margin trajectory. On April 14, Adobe launched Firefly AI Assistant, a natural language interface that executes multi-step workflows across Creative Cloud applications including Firefly, Photoshop, Premiere, Lightroom, Express and Illustrator.

Adobe’s moves come as software stocks face pressure from the rise of autonomous AI tools, with competition from startups such as Anthropic and platforms like OpenAI. Anthropic recently introduced Claude Design, an experimental feature for generating visuals such as prototypes, slide decks and one-page documents from its chatbot.