Claude AI now makes going through long PDFs much easier

When it comes to AI chatbots, perhaps the first names that come to mind are OpenAI’s ChatGPT, Google Gemini, and Microsoft Copilot. However, another name, one nearly as capable as the other three, is Anthropic’s Claude AI chatbot.

Now, we say nearly, because in addition to generating textual responses to your queries, the chatbot also could analyse and summarise PDFs and images, as well as answer questions about them. What threw Claude off, however, were images within PDF files, which it couldn’t process. That’s all set to change however, with Claude Sonnet 3.5. Read on.

PDF image processing for Claude: What you need to know

As part of the recently unveiled Claude 3.5 Sonnet AI-powered upgrade, Claude AI now gets the ability to analyse charts and other images within PDFs, to perform a deeper analysis. Once done, users can ask the chatbot questions about those specific images and it can provide them with relevant answers. The feature was long overdue, especially considering how the chatbot already had the ability to process images earlier.

ALSO READ: I tried Claude AI, and it’s now my go-to chatbot app

The new feature is aimed especially at enterprise users, whose typical everyday work revolves around a lot of number crunching, especially for those in sales or finance roles. The fact that Claude 3.5 Sonnet accepts PDFs as an input also brings it on par with NotebookLM, Google’s dedicated AI-powered research and note-taking tool.

Bear in mind that a PDF uploaded to Claude should have no more than 100 pages. Additionally, the PDF shouldn’t be more than 32MB in size, and neither should it be password-protected.

How to access the new PDF image processing feature on Claude?

The new PDF-processing capabilities on Claude 3.5 Sonnet are still in beta. Therefore, you will require a subscription to a Pro plan for accessing the same, whether on your smartphones, laptops or desktops. If you use a Pro plan, click on your account name at the bottom of the left sidebar, and select ‘Feature Preview’ from the subsequent menu. Here, turn the toggle on next to the ‘Visual PDFs’ option.

All you’ve got to do now is return to the main screen of the chatbot and start a chat. Click on the paperclip icon and choose a PDF from your device, and after the file’s thumbnail appears under the prompt, type and submit your query.