Linux Foundation creates DNS framework for AI agents

The Linux Foundation announced a new project called the Agent Name Service on Wednesday to establish identity and trust for AI agents. This framework provides a standardized way for organizations to verify the ownership and permissions of autonomous systems, ensuring that AI interactions remain secure and transparent as deployments scale.

Establishing a Trust Layer for Autonomous Systems

The Agent Name Service framework functions as a directory for the next generation of digital assistants. It allows systems and human users to confirm exactly who an agent represents. This verification process covers the permissions granted to the agent and checks if its operational history is authentic. The framework uses the existing Domain Name System as its foundation to achieve these goals.

DNS usually translates easy-to-read website names into technical internet addresses. The Linux Foundation uses this same logic for AI agents. ANS creates a naming and discovery layer that is easy for humans to understand. Enterprises can publish these agent identities using the domains they already own. This setup helps other systems discover what an agent can do before any interaction starts.

This method creates a federated system for discovery and verification. It avoids the need for a single company to control a central registry. Because it is decentralized, no single entity owns the data or the access points. Organizations maintain full control over the identities they create. This autonomy is vital for businesses that handle sensitive data or operate in competitive markets.

The framework also ensures that the code running an agent has not been changed without authorization. Trust is built by showing a clear trail of ownership. If an agent tries to perform a task, the receiving system can check the ANS record. This record confirms if the agent has the legal right to act on behalf of a specific company.

Addressing the Demand for Identity Control

Industry experts see a growing need for this type of identity management. As AI agents move into production, they often interact across different APIs and tools. These interactions happen across organizational boundaries without a consistent way to track accountability. Charlie Dai, a principal analyst at Forrester, notes that identity issues are already appearing in early deployments.

Regulated industries face specific challenges with agent-to-agent interactions. These sectors require strict proof of where an AI system came from and what it is allowed to do. Without a standard like ANS, auditing these interactions becomes nearly impossible. Companies need to know which agent took an action and if that action matched its intended design.

Gartner analysts describe agent identity as a critical operational gap. It is no longer just a theoretical concern for architects. It is a practical problem for managers who oversee daily operations. They must verify that an agent has the authority to spend money or access private files. ANS provides a control plane to manage these risks effectively.

Using DNS as the base provides a major advantage for quick adoption. Most companies already have the infrastructure to manage domains. They do not need to build entirely new systems to start using ANS. This makes the transition cheaper and faster for IT departments. Relying on familiar internet protocols reduces the learning curve for staff.

However, using DNS is not without its risks. The system was not originally built for high-level security tasks. Critics point out that DNS can be vulnerable to hijacking or spoofing. There are also concerns about how long it takes for records to update across the globe. These delays could cause inconsistencies in trust guarantees during fast-paced operations.

To fix these gaps, the Linux Foundation does not rely on DNS alone. The framework includes support for Decentralized Identifiers and Legal Entity Identifiers. These tools allow businesses to link AI agents to existing corporate identity systems. By combining these technologies, the framework offers a more complete verification model than DNS could provide on its own.

Navigating a Competitive Landscape of Standards

The Agent Name Service enters a market that is already filling with different standards. Various protocols currently exist to connect agents to tools or help them talk to each other. The Linux Foundation itself hosts multiple projects that deal with discovery and trust. This variety can sometimes lead to confusion for businesses looking for a single solution.

Existing Frameworks and Overlap

One similar project is DNS-AI Discovery, which helps agents advertise their skills across a network. Another project, led by Cisco and known as AGNTCY, provides a full stack for multi-agent systems. It handles messaging and observability alongside identity. The overlap between these projects suggests that the industry is still searching for the best path forward.

Some observers worry that having too many standards will cause fragmentation. If different companies use different systems, agents might not be able to talk to each other. This would defeat the purpose of creating a universal naming service. However, some analysts believe this competition is healthy for the market. It allows different ideas to be tested in real-world scenarios.

The Future of Standardization

Market experts suggest that we are currently in a discovery phase for AI standards. It is natural to see multiple frameworks competing to solve the same problems. Over time, these ideas will likely merge into a more unified system. For now, the focus is on finding which methods work best for large-scale enterprise needs.

Enterprises should stay informed about these developments but move carefully. It is important to wait for clearer guidance on how these different systems will work together. Treating any single project as the final answer might be premature. The goal is to build a foundation that is flexible enough to adapt as the technology changes.

Security experts recommend using ANS alongside other tools. These include identity and access management systems and AI gateways. Combining ANS with API security controls creates a defense-in-depth strategy. This approach ensures that even if one layer fails, other protections remain in place to stop unauthorized actions.

As the Linux Foundation moves forward with ANS, the focus remains on interoperability. The success of the project depends on how well it works with other industry tools. If ANS becomes a bridge between different platforms, it will provide the trust layer that modern AI deployment requires. This project marks a significant step toward making autonomous AI a reliable part of corporate infrastructure.