How Adobe's enhanced AI can transform document management, customer engagement

Two new AI assistants - for Adobe Acrobat and the newly upgraded Adobe Experience Platform - improve productivity and efficiency. However, the company must demonstrate the strength of its security measures to gain user trust.
Written by Jason Perlow, Senior Contributing Writer

Adobe's AI assistant as shown within Adobe Experience Cloud


Adobe has entered the generative AI arena -- alongside other tech giants --  to significantly enhance its user experiences. The company's latest advancements, particularly through the Adobe Acrobat AI Assistant and the newly upgraded Adobe Experience Platform (AEP) AI Assistant, are revolutionizing how users interact with documents and manage customer engagement.  

This move aligns Adobe with the broader industry trend of integrating generative AI and large language models (LLMs) into actual products. Notable examples include Microsoft's GPT-4-based CoPilot and Google's Workspace offerings powered by Deepmind Gemini. 

Also: ChatGPT vs. Microsoft Copilot vs. Gemini: Which is the best AI chatbot?

Adobe's integration of GenAI models also mirrors HubSpot's efforts, which incorporates similar technologies into its digital marketing platform that competes with Adobe in its Experience Manager and Experience Cloud offerings.

How Adobe leverages GenAI models

Adobe's approach to integrating AI involves leveraging the Microsoft Azure OpenAI Service (the basis for Redmond's own CoPilot and the hosting platform for ChatGPT) and other top-tier GenAI technologies. However, Adobe has not specified which additional third-party models are being used beyond GPT-4 and its own Firefly models for image generation. 

Also: Microsoft releases upgrades to Azure AI Speech at Build 2024

Adobe's model-agnostic strategy ensures that the AI Assistant delivers high-quality, secure experiences while maintaining stringent data security protocols. The company states that no customer document content is used to train these third-party models, aligning with the company's commitment to data security and AI ethics

Adobe protects data through encryption, restricts access to authorized personnel, and adheres to industry standards and regulations. Their stated protocol includes regular security audits, continuous monitoring, and anonymizing customer data for privacy. Third-party partners must also meet these stringent security standards, safeguarding customer data and preventing unauthorized access.

Generative AI tools for advertising campaigns

At the Adobe Summit in March 2024, Adobe introduced several new generative AI tools to transform how brands create and manage their ad campaigns. One of the key announcements was GenStudio, an AI-first application that centralizes all content needs in one place. GenStudio enables users to create content, access brand assets, view and track campaigns, and measure campaign performance. AI assists with finding images and generating variations using Adobe Firefly, Adobe's generative AI suite.

Another significant offering is Custom Models, which allows enterprises to train and customize models. It includes fine-tuning Adobe Firefly with their brand's assets to ensure generated content remains on-brand. GenStudio uses a feedback loop to analyze the performance of generated assets, providing insights that inform future generative AI prompts.

GenAI within PDFs and more


The Adobe AI Assistant within PDF Reader


Adobe Acrobat AI Assistant, introduced as a beta in February 2024 and fully available in April, has already made its mark on document management. This tool, integrated into Acrobat and Reader, offers a conversational interface that makes document interaction more dynamic. Users can ask questions about document content, quickly summarize long texts, and efficiently navigate complex documents. The assistant also generates and formats content for various uses, such as emails, presentations, blogs, and reports, while providing intelligent citations to ensure accuracy.

Also: 6 ways Apple can leapfrog OpenAI, Microsoft, and Google at WWDC 2024

The AI Assistant's ability to handle different document formats, including Word and PowerPoint, further enhances its utility. Its voice command feature, currently in beta on mobile, allows users to interact with their documents hands-free, adding another layer of convenience and accessibility. Additionally, users can create compelling content for various platforms, consolidating and formatting information effortlessly. The full range of features is available through an add-on subscription starting at $4.99 per month.

GenAI within Adobe Experience Cloud

Building on the success of the Acrobat AI Assistant, Adobe launched the AEP AI Assistant on June 6. This new assistant integrates advanced AI capabilities within Adobe Experience Cloud applications, such as Adobe Real-Time Customer Data Platform (CDP), Adobe Journey Optimizer (AJO), and Customer Journey Analytics (CJA). The AEP AI Assistant offers powerful tools for marketers and other professionals:

  • Content Creation: Automatically generates marketing assets like emails, web pages, and personalized campaign content, including text and design elements, ensuring a cohesive and professional look.
  • Customer Journey Management: This helps create, optimize, and manage customer journeys by leveraging AI to suggest the next best actions based on historical data and predictive analytics, enhancing customer experiences with more relevant and timely content.
  • Predictive Insights: Provides detailed predictive insights and recommendations, such as predicting customer behavior (e.g., likelihood of a purchase or engagement), allowing for more targeted marketing efforts.
  • Automated Task Management: Answers technical questions, automates routine tasks, and provides operational insights without the need for complex queries, reducing the workload on team members and allowing them to focus on strategic initiatives.
  • Data Activation and Integration: Activates data across various channels and integrates it into existing Adobe Experience Cloud applications workflows, ensuring all marketing efforts are data-driven and aligned with business goals.

The AEP AI Assistant is designed to work across Adobe workflows where customers can benefit. Each AI Assistant is unique, depending on the product, and powered by Adobe's generative experience models. These models include an Adobe base model (deep product knowledge), a custom brand model (based on a brand's own data, audiences, and campaigns), and support for a large language model that meets the customer's needs.

Future outlook

Integrating generative AI across Adobe's products has profound implications for users and the industry. Both AI assistants significantly enhance productivity by enabling users to perform tasks more efficiently. Knowledge workers can quickly summarize meeting transcripts, create study guides, and optimize marketing campaigns, while marketing teams can generate and refine content and customer journeys.

The technology behind these AI assistants integrates Adobe's AI and ML models with custom brand models and LLMs for natural language processing, offering high levels of customization and flexibility tailored to individual brand preferences. This allows users to leverage advanced AI capabilities finely tuned to their specific needs, driving better outcomes and more effective workflows.

Also: More political deepfakes exist than you think, according to this AI expert

Adobe intends to further improve its AI assistants by adding more features. These enhancements include AI-driven authoring, editing, formatting, and intelligent document collaboration. The goal is to simplify the creation of initial drafts and content editing and provide suggestions for content design and layout, making content creation more efficient and accessible. Additionally, generative AI will improve digital collaboration by analyzing feedback and comments, suggesting changes, and helping to resolve conflicting feedback, thereby streamlining the process of moving from draft to final document.

Adobe's advancements in generative AI tools are extremely significant. They represent an evolution in how we handle documents and a transformation in productivity and efficiency across various sectors. The ability to quickly generate, format, and extract information from documents can save countless hours for professionals, allowing them to focus on more strategic tasks. For marketers, predictive insights and automation capabilities can drive more effective campaigns and improve customer engagement.

But will customers take to it?

Adobe's generative AI advancements in Acrobat and AEP AI Assistants have huge potential to transform document management and customer engagement. By integrating powerful AI capabilities into its products, Adobe enables users to work smarter and more efficiently, unlocking new value from their digital documents and customer interactions. 

Also: How to use ChatGPT to create an app

However, for these features to gain widespread adoption, Adobe must rigorously document and prove the stringency and privacy of its security protocols. While Adobe has committed to not using customer content to train third-party models, the company must ensure no data leaks between customer interactions. Providing secure LLM instances on an enterprise basis could further enhance data protection and reassure customers about the integrity of their data.

Editorial standards