Could MCP supercharge the agentic AI revolution?

Agentic AI is frequently touted as the next wave of innovation set to transform how businesses interact with customers and suppliers. Gartner predicts that by 2029, the technology will resolve 80% of common customer service issues resulting in a 30% reduction in operational costs. But a key challenge for providers and vendors is connecting LLMs to the multiple data sources needed to provide truly autonomous agentic AI. Legacy systems combined with proprietary protocols make data sharing across internal and external networks a challenge.The Model Context Protocol (MCP) launched as an open standard by LLM developer Anthropic in November 2024 has the potential to solve some of these challenges. The company describes MCP as enabling developers to build secure, two-way connections between their data sources and AI-powered tools. As opposed to more reactive protocols for sharing data across systems such as the Language Server Protocol (LSP), MCP allows autonomous agents to make calls and decisions on sourcing appropriate data and models to deal with requests. Some have referred to MCP as the USB-C port for AI applications.Open innovationBeing an open standard gives MCP the chance to gain rapid adoption in the same way the HTTP web protocol did for kickstarting the internet revolution. Similarly, email became a killer application for driving internet adoption through the open SMTP standard.If MCP starts to gain traction among developers and vendors, then a virtuous circle of innovation will likely develop as it becomes de facto. The lack of competing standards in this space also increases the likelihood of success for MCP.Going mainstreamWhile there’s a rapidly growing community of developers building with MCP, recent announcements from industry giants such as Microsoft, Cloudflare, and OpenAI say they support it driving the standard into the mainstream. Microsoft has added support for MCP to its Copilot Studio platform while Cloudflare now allows customers to build and deploy remote MCP servers.These developments are significant as they remove the need for inhouse MCP servers, opening up deployment opportunities to enterprises without extensive IT resources.OpenAI’s support for MCP is particularly significant being one of Anthropic’s key competitors. It’s a recognition that common connectivity enablers are crucial if AI is to really deliver value to businesses. “We’re entering the protocol era of AI,” says Alexander Doria, co-founder of AI startup Pleias. “This is how agents will actually do things.”What lies aheadDespite the momentum behind MCP, potential hurdles could derail its adoption. Security, discoverability, and corporate capture are three of the biggest challenges.The potential for security breaches when AI agents make autonomous calls on remote databases and servers is obvious. Anthropic has built levels of security into the protocol in terms of authentication, where calls are made on external data sources and servers, ensuring security will be more complex. We can expect vendors and solutions providers such as Microsoft and Cloudflare to bake security into their offerings as competitive differentiators.The key value of MCP is bringing together multiple tools, LLMs, and data sources, allowing autonomous agents to provide answers and solutions to real-world problems. The ease of discoverability of these resources in near real time is another challenge. Google solved the problem of locating information on the web 25 years ago through its indexing and PageRank algorithm. As users flocked to the search engine, website owners optimized their content for greater visibility, bending much of the web to Google’s algorithm. MCP servers are at the heart of this agentic AI transformation, and various initiatives are underway to catalog and provide access to them. MCP.so currently lists and offers connections to over 4,800 MCP servers with the number growing daily.Another potential challenge lies with the MCP standard forking into a more proprietary format through corporate capture. Microsoft tried to colonize the web in the 1990s through the Internet Explorer browser and use of its VBScript and Jscript scripting languages. Although ultimately unsuccessful, it could have derailed the explosion of digital innovation of the last 30 years.The greater goodDespite these challenges, there’s a positive future for MCP. The dynamic developer communities sharing information and best practices provide a strong foundation for experimentation and innovation. The more recent entry of tech giants in providing solutions to extend MCP’s potential beyond the desktop and across external networks should encourage CIOs to explore the standard’s potential for their organization. “As the wider market adopts AI Agents, and the ability to interoperate with reduced or zero integration effort,” says Jeremy Gaerke, CTO of data pipeline observability and traceability platform, Pantomath, “enterprise SaaS customers will expect agentic interoperability as a default. MCP could be the answer.

Agentic AI is frequently touted as the next wave of innovation set to transform how businesses interact with customers and suppliers. Gartner predicts that by 2029, the technology will resolve 80% of common customer service issues resulting in a 30% reduction in operational costs. But a key challenge for providers and vendors is connecting LLMs to the multiple data sources needed to provide truly autonomous agentic AI.

Legacy systems combined with proprietary protocols make data sharing across internal and external networks a challenge. The Model Context Protocol (MCP) launched as an open standard by LLM developer Anthropic in November 2024 has the potential to solve some of these challenges. The company describes MCP as enabling developers to build secure, two-way connections between their data sources and AI-powered tools.

As opposed to more reactive protocols for sharing data across systems such as the Language Server Protocol (LSP), MCP allows autonomous agents to make calls and decisions on sourcing appropriate data and models to deal with requests. Some have referred to MCP as the USB-C port for AI applications. Open innovation Being an open standard gives MCP the chance to gain rapid adoption in the same way the HTTP web protocol did for kickstarting the internet revolution.

Similarly, email became a killer application for driving internet adoption through the open SMTP standard. If MCP starts to gain traction among developers and vendors, then a virtuous circle of innovation will likely develop as it becomes de facto. The lack of competing standards in this space also increases the likelihood of success for MCP.

Going mainstream While there’s a rapidly growing community of developers building with MCP, recent announcements from industry giants such as Microsoft, Cloudflare, and OpenAI say they support it driving the standard into the mainstream. Microsoft has added support for MCP to its Copilot Studio platform while Cloudflare now allows customers to build and deploy remote MCP servers. These developments are significant as they remove the need for inhouse MCP servers, opening up deployment opportunities to .

OpenAI’s support for MCP is particularly significant being one of Anthropic’s key competitors. It’s a recognition that common connectivity enablers are crucial if AI is to really deliver value to businesses. “We’re entering the protocol era of AI,” says Alexander Doria, co-founder of AI startup Pleias.

“This is how agents will actually do things.” What lies ahead Despite the momentum behind MCP, potential hurdles could derail its adoption. Security, discoverability, and corporate capture are three of the biggest challenges.

The potential for security breaches when AI agents make autonomous calls on remote databases and servers is obvious. Anthropic has built levels of security into the protocol in terms of authentication, where calls are made on external data sources and servers, ensuring security will be more complex. We can expect such as Microsoft and Cloudflare to bake security into their offerings as competitive differentiators.

The key value of MCP is bringing together multiple tools, LLMs, and data sources, allowing autonomous agents to provide answers and solutions to real-world problems. The ease of discoverability of these resources in near real time is another challenge. Google solved the problem of locating information on the web 25 years ago through its indexing and PageRank algorithm.

As users flocked to the search engine, website owners optimized their content for greater visibility, bending much of the web to Google’s algorithm. MCP servers are at the heart of this agentic AI transformation, and various initiatives are underway to catalog and provide access to them. MCP.

so currently lists and offers connections to over 4,800 MCP servers with the number growing daily. Another potential challenge lies with the MCP standard forking into a more proprietary format through corporate capture. Microsoft tried to colonize the web in the 1990s through the Internet Explorer browser and use of its VBScript and Jscript scripting languages.

Although ultimately unsuccessful, it could have derailed the explosion of digital innovation of the last 30 years. The greater good Despite these challenges, there’s a positive future for MCP. The dynamic developer communities sharing information and best practices provide a strong foundation for experimentation and innovation.

The more recent entry of tech giants in providing solutions to extend MCP’s potential beyond the desktop and across external networks should the standard’s potential for their organization. “As the wider market adopts AI Agents, and the ability to interoperate with reduced or zero integration effort,” says Jeremy Gaerke, CTO of data pipeline observability and traceability platform, Pantomath, “enterprise SaaS customers will expect agentic interoperability as a default. MCP could be the answer.

.