What challenges did Box face before AI implementation?

Democratize AI agents to business users while ensuring agents do not access data they shouldn't. Enable multi-agent workflows across enterprise systems.

What was the impact of the AI solution?

Secure, permission-scoped agent workflows across enterprise teams. Claims scalable AI capabilities and cost efficiency. Users can search, analyze, trigger workflows, and create documents from Box content.

Source

GCPEvidence: Low40/100

BOX uses Gemini Enterprise Agent Platform and BigQuery to enable customers to build AI agents on Box content security boundaries

Use case typeAgent orchestrationUpdated Jun 13, 2026

Box builds customizable AI and AI agents with Google Cloud to democratize AI into workers' hands. Box Agent and Box AI Studio let non-technical users create custom, purpose-built AI agents over Box content. Agents authenticate to a company's Box instance and operate within existing user permissions and security settings. Multi-agent systems use Agent2Agent (A2A) and Box MCP server to interact with other Gemini Enterprise agents. BigQuery is used for reporting and processing structured aspects of unstructured data.

Organization: Box
Industry: Tech & Comms
Location: United States
Published: June 2026

Reported outcomes

Strategic outcomes

New product / capabilityEnabled no-code custom AI agentsRisk & complianceKept agents within user permissionsEcosystem & partnershipsEnabled multi-agent interoperabilityBetter decisions & insightImproved structured data reporting

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Box
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Customer Story

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Agentic content management
2Enterprise search and workflow automation
3Multi-agent orchestration

Democratize AI agents to business users while ensuring agents do not access data they shouldn't.
Enable multi-agent workflows across enterprise systems.

No-code Box AI Studio and Box Agent for configuring AI agents.
Integration with Gemini Enterprise Agent Platform and Google models.
Permission-scoped access to Box files, folders, and Box Hubs.
A2A protocol and Box MCP server for interaction with other agents.
BigQuery for processing and reporting structured aspects.

Technologies

Gemini Enterprise Agent Platform BigQuery Gemini Box AI Studio Box Agent Agent2Agent (A2A) protocol Box MCP server

Secure, permission-scoped agent workflows across enterprise teams.
Claims scalable AI capabilities and cost efficiency.
Users can search, analyze, trigger workflows, and create documents from Box content.

Architecture

Box's Box Agent and Box AI Studio provide no-code configuration for custom AI agents that authenticate to each customer's Box instance and operate with existing permissions. The solution integrates Gemini Enterprise Agent Platform and Gemini models, uses Agent2Agent (A2A) protocol and the Box MCP server for multi-agent interoperability, and leverages BigQuery for reporting and processing structured aspects of unstructured content.

Sources & evidence1

Evidence: Low40/100Evidence strength