Amidst a vast ocean of unstructured corporate forms, records, and documents lies valuable information that, if you can find it, can drive business processes in legal, sales, marketing, human resources, and other scenarios.
In many organizations, the task of structuring these files is typically left to content managers, business analysts, data scientists, and similar professionals. Box has released Box Extract, a generative AI agent. This can change this by allowing employees closer to the front lines, those who are the first to discover use cases for a particular data within a workflow, to discover and tag data.
This agent can also streamline processes such as data discovery, add proper structure when launching new document types, and update analytical dashboards for professional users.
Box Extract uses large language models such as Anthropic Claude Opus 4.5, Google Gemini 3, and OpenAI GPT 5.2 to transform plain language declarations into document extraction rules. These rules can be searched within Box or integrated with workflows in other apps such as Microsoft Office, Oracle Fusion Cloud HCM, Salesforce, ServiceNow, Slack, and Workday. AI metadata agents are a good first step toward enabling customers to classify unstructured data, said Alan Peltz Sharp, founder of independent technology research firm Deep Analysis.
“You can store things well, but you want to get value out of it.” [unstructured data]”Box users who don’t use it as a storage system, but actually use it in their day-to-day business, will get a lot of value from Box Extract,” said Pelz-Sharpe.
Peltzsharp added that an opportunity for Box could be custom versions of Box Extract that address common customer areas such as insurance, healthcare, and finance. Each industry has its own data model and specific needs. Box could reduce friction for large customers by creating accelerators that take into account complexities such as medical and auto insurance claims.
“There’s a huge growth area. People are looking for something really specific that understands their niche,” Peltzsharp said. “Box is very horizontal by design. But growth comes from building at least part of the company into these rich verticals.”
Box Extract was previewed at the BoxWorks User Conference last fall. This grew in part from technology acquired from Alphamoon in 2024. Box customers will now be able to manage security policies to prevent employees from building Box Extract agents that gain unauthorized access to protected information such as financial, health, and human resources records.
This tool, along with Box’s Generated AI content creation and summarization tools, represents the company’s tactical application of AI to perform specific operations, as opposed to the large-scale AI platform approaches of other vendors such as ServiceNow, Salesforce, and Microsoft.
“If you try to do too many crazy things with AI, if you try to force it to do too much, it’s often going to struggle just like humans,” said Ben Kass, Box’s chief technology officer. “But if you have a set of practical, difficult problems, problems that you couldn’t solve before, that’s a great way to make an AI project successful.”
Don Fluckinger is a senior news writer at Informa TechTarget. He is responsible for customer experience, digital experience management, and end-user computing. Any tips? send him an email.
