Docker Forms Alliance to Tame AI Agents

Docker has co-founded the Blueprint Alliance to create standards for the coming explosion of AI agents, which Gartner predicts will hit 150,000 per company.

The number of AI agents inside large companies is set to skyrocket, but the tools to manage them are lagging far behind. Citing a Gartner prediction that the average Fortune 500 company will go from using 15 AI agents today to over 150,000 by 2028, Docker announced it has co-founded the Blueprint Alliance. In a recent blog post, the company described the initiative as a necessary collaboration to address the immense security, governance, and operational challenges posed by this explosion. With most organizations feeling unprepared for the risks of today's AI, let alone tomorrow's, the alliance aims to create a standardized framework for building, deploying, and managing AI agents at scale. The core problem is that without common rules, companies face a chaotic and dangerous future where thousands of autonomous systems operate without clear oversight, creating a massive attack surface and unpredictable behavior.

The Blueprint Alliance intends to establish a set of common standards and best practices for the entire lifecycle of AI agents. While the initial announcement is high-level, Docker's involvement points to a solution rooted in proven DevOps principles. The goal is to define a consistent way to package agents, likely using containers, to ensure they run reliably across different environments. This approach would also integrate security scanning and software supply chain management, tools that are already central to the modern development process. By creating a shared “blueprint,” the alliance hopes to provide developers with a trusted, repeatable pattern for building agents that are secure and observable by default. This would transform the current ad-hoc creation of agents into a structured engineering discipline, making them easier to test, monitor, and govern once they are deployed in production environments.

This effort to standardize a rapidly emerging technology follows a familiar pattern in software history. The rise of AI agents mirrors the early days of web applications and microservices, which were initially built in a fragmented, chaotic manner before frameworks and platforms brought order. Docker itself played a pivotal role in the last major shift by providing a standard unit of deployment—the container—that solved the “it works on my machine” problem and enabled the microservices revolution. The Blueprint Alliance represents a similar attempt to get ahead of the curve for the “agentic era.” The industry is recognizing that for AI agents to become reliable, enterprise-grade tools rather than risky science projects, they need a robust underlying infrastructure and a common set of rules. This initiative is a bet that collaboration on foundational standards is the only way to unlock the full potential of agentic AI without succumbing to the complexity and risk.

For developers, CTOs, and security leaders, the formation of the Blueprint Alliance is a significant signal to begin formalizing their own internal strategies for AI agent governance. While the alliance's work is just beginning, its existence validates the growing concern around unmanaged AI adoption. Companies should start evaluating how they will track, secure, and manage the agents their teams are inevitably building and using. The immediate takeaway is not to adopt a specific tool, but to recognize the problem and start the conversation about internal policies. In the coming months, the industry should watch for the first draft specifications released by the alliance and, crucially, which other major cloud providers, AI companies, and enterprise software vendors join the effort. The breadth of its membership will ultimately determine whether these blueprints become a true industry standard or just one of several competing approaches.

The proliferation of AI agents without standards creates a fragmented and insecure landscape for developers and ops teams. This alliance aims to establish common patterns for agent development, security, and deployment, potentially preventing a future of unmanageable technical debt and security vulnerabilities across the software supply chain.

Unmanaged AI agent adoption poses significant security and compliance risks, threatening to halt innovation or expose companies to breaches. A standardized ecosystem, like the one proposed by the Blueprint Alliance, allows businesses to safely scale AI initiatives, ensure governance, and maintain a competitive edge without sacrificing security.