Box lifts user query cap for Box AI, announces upcoming GPT-4o integration

Box cloud storage logo and branding pictured at the BoxWorks 2015 Conference at the Moscone Center in San Francisco, California, US.
(Image credit: Getty Images)

Cloud-based content management specialist Box has unveiled fresh enhancements to its Box AI offering, including unlimited user queries for enterprise users and upcoming integration with OpenAI’s GPT-4o.

Enterprise Plus customers can now take advantage of unlimited end user queries in Box AI for Hubs, Documents, and Notes, while GPT-4o integration, support for image and spreadsheet files, and the Box AI for Metadata API are expected to be available later this year.

Launched in 2023, the Box AI capabilities suite integrates advanced AI models to Box’s Content Cloud – a platform leveraged by more than 115,000 enterprises to secure, store, and collaborate on critical business content.

Built on the firm’s platform-neutral framework, the offering connects with popular large language models (LLMs) including Microsoft’s Azure OpenAI service and Google Cloud’s Vertex AI.

As demand for Box AI has increased since its launch, Box said is now removing its query caps so enterprise users have unlimited access to enterprise-grade AI capabilities. Previously, the plan offered users 20 queries per month, with 2,000 additional queries available on a company level.

“The improvements we’ve made to the Box AI platform – combined with the decreasing cost of AI over the past few months – have driven tremendous efficiencies and, as a result, we are removing our query limits in Box AI,” explained Ben Kus, chief technology officer at Box.

“The combination of AI and unstructured data represents the biggest untapped opportunity in enterprise IT, and we want Box customers to be on the leading edge of innovation with Box AI as a core part of the Box platform.”

All the new Box integrations and features

In addition to unlimited user queries, Box is also introducing support for OpenAI’s new GPT-4o later this year. 

The firm said the model offers advanced capabilities that will help Box Hubs improved answers and support for more than 20 additional languages.

Box Hubs users will be able to leverage enhanced text-related functionality that includes multimodal understanding, ethical and sensitive content handling, as well as task-specific adaptation.

File type support

With support for natural language queries on image file formats, customers will also soon be able to ask specific questions about images stored across the Box platform, with additional support to include querying on spreadsheets.

It means users will be able to run precise data analysis in spreadsheets – a function Box said will simplify interactions with large structured datasets and drive insight generation.

Metadata API

Additionally, Box is introducing the Box AI for Metadata API which will enable developers to integrate Box AI with custom applications to automatically extract key information and data from documents at scale. 

Currently in beta for customers on Enterprise Plus plans, users can pair the API with Box’s workflow automation tools to automate processes based on file metadata, extract key fields from unstructured content, as well as save information to external applications such as Salesforce.

Pricing for the Box AI for Metadata API will be revealed closer to general availability, Box said, along with pricing for other platform API calls and end user metadata queries within the core Box application.

Commenting on the additions, Aaron Levie, co-founder and CEO of Box, said Box AI’s expanded functionality will help enterprise customers drive further efficiencies across their businesses.


“Since Box AI became generally available earlier this year, we’ve seen significant interest from customers to upgrade to Enterprise Plus plans so they can benefit from the power of AI with their content,” he said.

“Today’s announcements will make it even easier for customers to roll out Box AI across their enterprise while also delivering new innovation so they will be able to get additional value from Box AI.”

Daniel Todd

Dan is a freelance writer and regular contributor to ChannelPro, covering the latest news stories across the IT, technology, and channel landscapes. Topics regularly cover cloud technologies, cyber security, software and operating system guides, and the latest mergers and acquisitions.

A journalism graduate from Leeds Beckett University, he combines a passion for the written word with a keen interest in the latest technology and its influence in an increasingly connected world.

He started writing for ChannelPro back in 2016, focusing on a mixture of news and technology guides, before becoming a regular contributor to ITPro. Elsewhere, he has previously written news and features across a range of other topics, including sport, music, and general news.