Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Microsoft MXC gives AI agents a controlled environment to run code, with policies that define which files they can read, where they can write, and what they can access. This demo shows MXC running natively on Windows, connected to governed business content in Box through Box Drive. Only approved... files are staged into MXC before the agent runs. Box Drive makes the content accessible through the desktop filesystem. When the work is finished, results return to Box for team review. MXC enforces what the agent can do. Box provides secure, governed content management and brings the output back into the team's workflow. Together, they connect useful agent work with enterprise controls. Demo below ↓show more

Box

80,249 subscribers

143,801 views • 2 days ago •via X (Twitter)

News & Politics Education Science & Technology

Anya Rossi• Live Now

Private livecam show

3 Comments

why2 days ago

@Windows The file-level policies are the interesting part here. Agents need clear boundaries around what they can read and where they can write.

Sujal2 days ago

@Windows file level read and write policies are the only sane way to hand an agent a real filesystem

Blanc2 days ago

@Windows staging is point-in-time: the agent reads staged bytes while the box source keeps moving. without a hash at stage-in and a re-check at return, the result diff is fiction.

Related Videos

AI assistants are moving from "answer my question" to "do the work." But they are only as useful as the enterprise content they have access to. Our demo shows what it looks like when Microsoft Copilot Cowork is grounded in Box. Your governed content powering multi-step agentic workflows, without ever leaving Microsoft 365.

AI assistants are moving from "answer my question" to "do the work." But they are only as useful as the enterprise content they have access to. Our demo shows what it looks like when Microsoft Copilot Cowork is grounded in Box. Your governed content powering multi-step agentic workflows, without ever leaving Microsoft 365.

Box

724,144 views • 4 months ago

AI agents need more than intelligence. They need trusted business context. With Google Gemini Managed Agents now supporting remote MCP servers, developers can connect Gemini agents to governed content in Box, generate outputs from real enterprise files, and save the work back where teams already collaborate. Box MCP Server + Gemini API in action.👇

AI agents need more than intelligence. They need trusted business context. With Google Gemini Managed Agents now supporting remote MCP servers, developers can connect Gemini agents to governed content in Box, generate outputs from real enterprise files, and save the work back where teams already collaborate. Box MCP Server + Gemini API in action.👇

Box

458,101 views • 3 months ago

OpenAI and Box are bringing your enterprise content directly into ChatGPT. As the file system for AI, Box provides headless access to enterprise context in ChatGPT while keeping the content managed, governed and protected by your existing Box permissions. Demo ↓

OpenAI and Box are bringing your enterprise content directly into ChatGPT. As the file system for AI, Box provides headless access to enterprise context in ChatGPT while keeping the content managed, governed and protected by your existing Box permissions. Demo ↓

Box

3,478,050 views • 29 days ago

I’ve been waiting to post this! The Box Agent skill now ships pre-installed with Nous Research Hermes agent. Box gives Hermes the content and context to become a real collaborator at work. Update Hermes, prompt `connect to Box`, and your agent can now search, read, and write across Box with Box AI. Here’s a quick demo 👇

I’ve been waiting to post this! The Box Agent skill now ships pre-installed with Nous Research Hermes agent. Box gives Hermes the content and context to become a real collaborator at work. Update Hermes, prompt `connect to Box`, and your agent can now search, read, and write across Box with Box AI. Here’s a quick demo 👇

Chris Kim

49,078 views • 1 month ago

🆕 OpenAI Developers Agents API gives developers a hosted sandbox where agents run tools, coordinate work, and handle complex tasks. Box Mount brings enterprise content directly into that sandbox as files the agent can read, reason across, and produce new work from, using normal shell commands and file paths while Box Mount handles two-way sync automatically. In our demo, a lead agent mounts a deal room from Box, reads five source files, launches three specialist agents in parallel, and writes four reports back through Box Mount with an initial verdict. When a customer revises the MSA in Box, Box Mount syncs the new version into the same sandbox and the agents reassess automatically, refreshing all four reports while preserving the revised MSA as version two in Box. The result is a shared workspace where people and agents work on the same governed Box content without building custom file-transfer logic, with permissions, versions, and audit history preserved throughout. Box Mount is in private preview. Watch👇

🆕 OpenAI Developers Agents API gives developers a hosted sandbox where agents run tools, coordinate work, and handle complex tasks. Box Mount brings enterprise content directly into that sandbox as files the agent can read, reason across, and produce new work from, using normal shell commands and file paths while Box Mount handles two-way sync automatically. In our demo, a lead agent mounts a deal room from Box, reads five source files, launches three specialist agents in parallel, and writes four reports back through Box Mount with an initial verdict. When a customer revises the MSA in Box, Box Mount syncs the new version into the same sandbox and the agents reassess automatically, refreshing all four reports while preserving the revised MSA as version two in Box. The result is a shared workspace where people and agents work on the same governed Box content without building custom file-transfer logic, with permissions, versions, and audit history preserved throughout. Box Mount is in private preview. Watch👇

Box

229,908 views • 28 days ago

Box is now available in Muse, from Meta! Your personal AI agent can now access governed enterprise content from Box, with permissions intact. One secure place where everything works together. Connect it in Muse today.

Box is now available in Muse, from Meta! Your personal AI agent can now access governed enterprise content from Box, with permissions intact. One secure place where everything works together. Connect it in Muse today.

Box

1,263,222 views • 10 days ago

AI tools are only as useful as the content they can reach. The Box Connector is now live for all paying Grok users, giving Grok Business secure, permissions-aware access to the enterprise content already living in Box. Watch here. 👇

AI tools are only as useful as the content they can reach. The Box Connector is now live for all paying Grok users, giving Grok Business secure, permissions-aware access to the enterprise content already living in Box. Watch here. 👇

Box

4,673,081 views • 4 months ago

🚀Vercel Sandbox gives AI agents an on-demand Linux microVM with its own filesystem, network, and process space. Box Mount runs inside that sandbox and maps a Box folder to a standard POSIX path, with 2-way sync and Box permissions and governance applying automatically. In this demo, the agent analyzes 12 security questionnaire items against 3 approved policy documents. 10 answers pass with sourced evidence. 2 are flagged for human review because the approved docs don't support those claims. The finished Excel workbook and review list sync back to Box through Box Mount. Vercel Sandbox is the isolated execution layer, and Box Mount is the governed content layer. The result lands in Box, ready for human review.

🚀Vercel Sandbox gives AI agents an on-demand Linux microVM with its own filesystem, network, and process space. Box Mount runs inside that sandbox and maps a Box folder to a standard POSIX path, with 2-way sync and Box permissions and governance applying automatically. In this demo, the agent analyzes 12 security questionnaire items against 3 approved policy documents. 10 answers pass with sourced evidence. 2 are flagged for human review because the approved docs don't support those claims. The finished Excel workbook and review list sync back to Box through Box Mount. Vercel Sandbox is the isolated execution layer, and Box Mount is the governed content layer. The result lands in Box, ready for human review.

Box

96,586 views • 1 day ago

AI agents can read any file the user has access to. That is a much larger surface than most security teams realize. Classification-based access policies in Box now give admins read access control for AI agents. Restrict which agents can read which content based on Box Shield classifications. Enforced automatically across every agent, regardless of which AI tool is making the request. This demo shows it in action.

AI agents can read any file the user has access to. That is a much larger surface than most security teams realize. Classification-based access policies in Box now give admins read access control for AI agents. Restrict which agents can read which content based on Box Shield classifications. Enforced automatically across every agent, regardless of which AI tool is making the request. This demo shows it in action.

Box

105,605 views • 2 months ago

Agents are moving from copilots to operators. But operational agents need something simple and powerful: a real filesystem to operate in. Not just chat. Not just APIs. A governed content environment where they can read, write, route, and collaborate. In this demo, OpenClaw🦞 operates inside Box — scoped to a specific intake folder — turning contracts into structured workflow within a secure, permission-aware system. This is how agents begin to operationalize enterprise work. Dive in to get started:

Agents are moving from copilots to operators. But operational agents need something simple and powerful: a real filesystem to operate in. Not just chat. Not just APIs. A governed content environment where they can read, write, route, and collaborate. In this demo, OpenClaw🦞 operates inside Box — scoped to a specific intake folder — turning contracts into structured workflow within a secure, permission-aware system. This is how agents begin to operationalize enterprise work. Dive in to get started:

Box

297,884 views • 7 months ago

AI tools will keep changing, but content policy needs to travel with the content. See how Box Shield applies classification-based access controls to what AI agents can actually read, not just what they can download. 👇

AI tools will keep changing, but content policy needs to travel with the content. See how Box Shield applies classification-based access controls to what AI agents can actually read, not just what they can download. 👇

Box

100,744 views • 25 days ago

Today we are announcing Box Agent Security and Governance. AI agents are reading files, moving content, and triggering actions across your most sensitive data at machine speed. Most security tools were not designed for this. Box closes the gap at the content layer: · Prompt injection detection before it reaches the model · Agent guardrails on what agents can and cannot do · Classification-based access policies extending Box Shield to AI · Audit trails for every agent session No additional software. No per-vendor configuration.

Today we are announcing Box Agent Security and Governance. AI agents are reading files, moving content, and triggering actions across your most sensitive data at machine speed. Most security tools were not designed for this. Box closes the gap at the content layer: · Prompt injection detection before it reaches the model · Agent guardrails on what agents can and cannot do · Classification-based access policies extending Box Shield to AI · Audit trails for every agent session No additional software. No per-vendor configuration.

Box

749,502 views • 2 months ago

🆕SpaceXAI just launched Grok Bot, a personal AI assistant built for complex knowledge work. Connected to Box through the Box MCP server, Grok Bot can work across your governed enterprise content immediately, across any workflow and any industry. In our demo it tackles a credit committee review. Works across credit agreements, borrower financials, diligence materials, collateral documentation, and risk assessments. Reconciles key terms. Flags credit risks and inconsistencies. Produces a source-cited brief. Saves it back to Box with structured deal metadata applied. All your content. Any AI agent. Watch here. 👇

🆕SpaceXAI just launched Grok Bot, a personal AI assistant built for complex knowledge work. Connected to Box through the Box MCP server, Grok Bot can work across your governed enterprise content immediately, across any workflow and any industry. In our demo it tackles a credit committee review. Works across credit agreements, borrower financials, diligence materials, collateral documentation, and risk assessments. Reconciles key terms. Flags credit risks and inconsistencies. Produces a source-cited brief. Saves it back to Box with structured deal metadata applied. All your content. Any AI agent. Watch here. 👇

Box

1,399,288 views • 1 month ago

Box and OpenAI are partnering to bring governed enterprise content to AI-powered legal work with Astra for Law and ChatGPT. Legal skills built on Box and ChatGPT help law firms turn their content into practical, secure workflows without weakening the controls around it. In this demo, a legal professional uses the new Outside Counsel Guidelines skill to review an incoming set of OCGs directly from Box. ChatGPT flags associated risks and leaves comments for human review, all without the content leaving Box.

Box and OpenAI are partnering to bring governed enterprise content to AI-powered legal work with Astra for Law and ChatGPT. Legal skills built on Box and ChatGPT help law firms turn their content into practical, secure workflows without weakening the controls around it. In this demo, a legal professional uses the new Outside Counsel Guidelines skill to review an incoming set of OCGs directly from Box. ChatGPT flags associated risks and leaves comments for human review, all without the content leaving Box.

Box

959,248 views • 22 days ago

AI agents work better when they bring in context step by step. With Box Agent skills and progressive disclosure of LangChain skills, developers can quickly build AI agents that load the right skill at the right time, stay grounded in Box content, and keep access tightly scoped. Less context bloat. More control. Faster enterprise AI development.

AI agents work better when they bring in context step by step. With Box Agent skills and progressive disclosure of LangChain skills, developers can quickly build AI agents that load the right skill at the right time, stay grounded in Box content, and keep access tightly scoped. Less context bloat. More control. Faster enterprise AI development.

Box

627,810 views • 5 months ago

Introducing the Box HTML editor. Create, edit, preview, and collaborate on HTML files directly in Box, no additional downloads, no external tools, no re-uploads. Your HTML files can now operate in the same governed space as the rest of your enterprise content, with seamless Box AI access built in from the start.

Introducing the Box HTML editor. Create, edit, preview, and collaborate on HTML files directly in Box, no additional downloads, no external tools, no re-uploads. Your HTML files can now operate in the same governed space as the rest of your enterprise content, with seamless Box AI access built in from the start.

Box

3,405,237 views • 3 months ago

AI agents need a structured place to do their work and access relevant content. A filesystem provides a clear, navigable way to manage that context. Here’s how Box can back that abstraction in LangChain Deep Agents — preserving the simplicity of the filesystem interface while adding enterprise-grade security, permission controls, version history, and collaboration at the storage layer. Explore our implementation:

AI agents need a structured place to do their work and access relevant content. A filesystem provides a clear, navigable way to manage that context. Here’s how Box can back that abstraction in LangChain Deep Agents — preserving the simplicity of the filesystem interface while adding enterprise-grade security, permission controls, version history, and collaboration at the storage layer. Explore our implementation:

Box

1,018,021 views • 7 months ago

AI agents are only as powerful as the content they can access. Anthropic announced Dispatch in Claude Cowork — assign tasks from your phone, let Claude navigate your computer, get complex work done autonomously. But without a file system built for AI, even the most capable agent is working blind. Connect Claude Cowork to Box and your files become the foundation it reasons from. Is your content infrastructure ready for AI agents?

AI agents are only as powerful as the content they can access. Anthropic announced Dispatch in Claude Cowork — assign tasks from your phone, let Claude navigate your computer, get complex work done autonomously. But without a file system built for AI, even the most capable agent is working blind. Connect Claude Cowork to Box and your files become the foundation it reasons from. Is your content infrastructure ready for AI agents?

Box

432,932 views • 6 months ago

Most teams governing AI agents focus on what they can access. The harder problem is what they can do with it. Box MCP Server admin controls let admins govern tool access enterprise-wide or per AI integration, set read-only or read-and-write permissions, and configure down to the individual tool level. When a tool is disabled, agents can no longer discover or invoke it. Enforced at runtime, not just at setup.

Most teams governing AI agents focus on what they can access. The harder problem is what they can do with it. Box MCP Server admin controls let admins govern tool access enterprise-wide or per AI integration, set read-only or read-and-write permissions, and configure down to the individual tool level. When a tool is disabled, agents can no longer discover or invoke it. Enforced at runtime, not just at setup.

Box

475,096 views • 1 month ago

With OpenAI’s AgentKit Agent Builder and the Box MCP server, you can build agents conversationally to automate complex workflows, while your data stays safe in Box. Ensuring your agents have the enterprise context they need to be effective and reliable.

With OpenAI’s AgentKit Agent Builder and the Box MCP server, you can build agents conversationally to automate complex workflows, while your data stays safe in Box. Ensuring your agents have the enterprise context they need to be effective and reliable.

Box

161,526 views • 1 year ago