Adobe’s AI Assistant Can Summarize PDFs, PowerPoints for You

Adobe is rolling out a new AI Assistant tool in beta for its Acrobat and Reader applications, allowing users to ask the chatbot questions about their documents or summarize main points, the software firm announced Tuesday.

Unlike most other generative AI models, Adobe’s AI assistant comes with a “custom attribution engine,” allowing the AI to actually cite its sources, according to the company. Adobe’s assistant is also able to hyperlink to different sections of text in its responses, giving users another easy way to verify the AI’s findings.

The Adobe assistant can work with PDFs as well as Microsoft Word docs and PowerPoints imported into Acrobat or Reader. Meeting transcripts can also be summarized, for example.

Adobe has released a series of short videos suggesting different use cases for the tool, from fine-tuning resumes to creating bullet points summarizing earnings calls to helping students study for tests.

While Adobe did not share what data its AI assistant was trained on, the company claims that no customer data parsed by the tool will be passed on to third parties or be used by Adobe itself.

“AI Assistant features in Reader and Acrobat are governed by data security protocols and no customer document content is stored or used for training AI Assistant without their consent,” Adobe said in its announcement.

While Adobe won’t charge users extra to get access to its AI assistant, the only free way to test the tool will be through Adobe’s Acrobat Reader desktop app, which will get the AI Assistant sometime in “the next few weeks.”

The AI assistant will also be added to Acrobat Standard and Pro Individual subscriptions, which cost $13 a month and $20 a month, respectively. The tool will only be available in English to start, with more language support planned in the future.

The company behind the Photoshop and Premiere Pro apps previously launched AI-powered text-based video editing. It also released an AI-powered image generator called Adobe Firefly, which can generate image elements from text prompts and create new brush textures, to name a few of its features.