Like all transformative technologies, such as email in the workplace or even calculators in classrooms, becoming mainstream takes time. We can think about the rise of AI agents in the workforce and the adoption of Anthropic's Model Context Protocol (MCP) -- a new standard for linking AI assistants directly to the systems where data lives -- as the latest trends in this cycle.
The term AI agent has gained popularity only in the past year, highlighting just how new agents are. Many enterprises are experimenting with AI agents, but few have fully integrated them into everyday workflows. This is partly because, like most new technologies, agents require improvements to become truly useful for users.
A major obstacle to AI adoption is connecting AI systems to the right enterprise tools and data in a secure, consistent way. As a result, AI agents are promising, but not quite applicable across every workflow.
Related:Will the music stop for AI's funding dance?
This is quickly changing. It seems like every week brings a new model update or improved interoperability between agents and the context they need to perform accurately. New developments are pushing the capabilities of AI agents to the next level, largely thanks to MCP.
Enterprises adopting MCP are creating a more reliable way for AI systems to access the data they need. You can think of MCP like a well-designed highway for AI and data. Instead of each company building its own disconnected roads, MCP provides a standardized route for data to move quickly and securely to the agents. As more companies use MCP servers to connect with agents from other platforms, agents will become more helpful in real-world applications.
You can think of MCP like a well-designed highway for AI and data.
Access to context across platforms: AI agents are only as useful as the context they can access. By standardizing how AI systems connect to data, MCP allows agents to work together across platforms, enabling context-aware applications.
Imagine a sales rep prepping for a customer call. Instead of logging into multiple systems, an AI agent powered by MCP can instantly pull the latest CRM updates, fetch supporting documents, and even coordinate workflows across apps like ServiceNow or Snowflake. With a secure API call through MCP, the agent gets exactly the context it needs to deliver relevant insights.
Compounding AI ecosystem value: MCP is emerging as the new rulebook for enterprise AI, and its impact grows exponentially as each company adopts it. The more companies that adopt the protocol, the more interoperable AI agents become, creating a virtuous cycle.
Enterprise-grade security: With MCP, AI models don't need direct access to every system or database, they just need to know which MCP servers are available. Each server enforces strict access controls, ensuring that AI agents can interact with only the data and actions they are authorized to use. This reduces the risk of unauthorized access or data leaks while maintaining its context-aware functionality.
Related:The hidden high cost of training AI on AI
As MCP adoption spreads, AI agents will progress. Each new implementation strengthens the ecosystem and provides a huge value-add for customers who can use AI agents across platforms for their personal workflows without worrying about security leaks. The more companies embrace MCP, the closer we get to a future where AI agents are fully integrated partners in everyday work.
Box
Ben Kus is CTO at Box, where he leads technology and AI strategy to help enterprises securely unlock insights from their unstructured data. Ben's career spans engineering, product leadership and startup innovation -- including co-founding Subspace (acquired by Box). He was an early employee at BigFix (acquired by IBM), where he later served as chief architect of mobile security. Ben holds a degree in computer science from University of California, Berkeley.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。