The Internet's New ID Card: Securing the Coming Age of AI Agents

SAN FRANCISCO, CA – June 23, 2026 – The Linux Foundation, a quiet giant behind much of the internet's open-source plumbing, today announced a plan to give the burgeoning world of artificial intelligence agents something they desperately need: a passport. The new initiative, called the Agent Name Service (ANS), aims to create a trusted identity system for the autonomous programs that are rapidly beginning to act on our behalf online. Instead of building a new digital country from scratch, the Foundation is doing something far more clever: it's renovating the internet's 40-year-old, battle-tested phonebook, the Domain Name System (DNS).

As a market analyst, I learned that the most profound shifts often aren't the flashiest; they're the foundational ones. And this move is squarely in that category. While we're dazzled by what AI can do, the real story is about building the guardrails to ensure it acts safely and accountably. ANS is a direct response to a growing anxiety in boardrooms and server rooms alike: how do we know who, or what, we're dealing with when an AI agent comes knocking?

The Rise of 'Shadow AI'

For most of us, AI agents are still a novelty. But in the corporate world, they are already being put to work. According to recent World Economic Forum data, a staggering 82% of executives plan to deploy AI agents in the next one to three years. The problem is, they're moving faster than the infrastructure built to manage them.

Industry analysts at firms like Gartner and Forrester have been sounding the alarm for months. Their reports describe a world where AI agents are already signing into systems, triggering workflows, and accessing sensitive data, often using identity systems designed for humans. This creates what security experts call "shadow AI"—autonomous agents operating without clear identity, authorization, or audit trails. For Chief Information Security Officers, this is a nightmare scenario, with top risks for 2026 including "rogue AI agents" and "AI identity sprawl."

Without a reliable way to verify an agent's identity, organizations can't answer basic questions: Who does this agent work for? What is it allowed to do? Can I trust that it hasn't been tampered with? This uncertainty is a major blocker to moving agents from promising experiments to production-scale systems. The Linux Foundation's ANS project is a direct attempt to solve this foundational trust problem before it spirals out of control.

An Old Solution for a New Problem

The elegance of the Agent Name Service lies not in its novelty, but in its pragmatism. Rather than inventing a new, proprietary identity network, ANS anchors agent identity directly to the DNS—the same global, decentralized system that turns names like linuxfoundation.org into network addresses. It's one of the most resilient and scaled pieces of infrastructure ever built, handling over 100 million queries per second worldwide.

"AI agents will increasingly operate across enterprises, platforms, and digital services, which makes trusted identity infrastructure a foundational requirement," said Jim Zemlin, CEO of the Linux Foundation, in the announcement. "By building on DNS and open standards, ANS creates a scalable and interoperable framework for verified agent communication."

This approach means that an organization can use the domain names it already owns to issue identities to its agents. An agent operating on behalf of a company could have a verifiable name tied to that company's domain, cryptographically secured by existing DNS security extensions (DNSSEC). This allows anyone or any system to verify who the agent represents, what permissions it holds, and whether its operational history is authentic.

This sentiment is echoed across the industry. "The global success of the Internet is rooted in shared, open principles, and its next phase driven by autonomous AI agents must be as well," commented Dane Knecht, CTO of Cloudflare. Wei Chen, Chief Legal Officer at Infoblox, put it even more directly: "That is the most efficient and responsible path forward: one shared specification, built on DNS, open to every organization and beholden to none."

A Battle for the Soul of the Agentic Web

This move by the Linux Foundation is about more than just technology; it's a strategic play in the battle for the future of the internet. Every major technological shift forces a choice between open, interoperable ecosystems and proprietary, centralized "walled gardens." The world of AI agents is no exception.

Major technology platforms are already developing their own agent identity solutions, which, while powerful, risk locking customers into a single vendor's ecosystem. If every platform has its own proprietary passport system, the dream of agents seamlessly interacting across the open web dies. It creates friction, increases costs, and stifles innovation.

ANS represents the open alternative. "The success of the internet didn't come from proprietary systems – it came from open standards, shared infrastructure, and an ecosystem committed to working together," noted Jared Sine, Chief Strategy and Legal Officer of GoDaddy. By championing an open standard under the neutral governance of the Linux Foundation, the project aims to ensure the "agentic web" doesn't become a collection of disconnected islands.

This is why participation from companies like Cisco is so significant. "Enterprises want to adopt these systems, but they cannot take on unbounded risk or build a new integration for every vendor," said Nathan Jokel, Senior Vice President of Corporate Strategy at Cisco. "That is why Cisco is contributing to open standards efforts... because the value of this technology will only compound when the ecosystem is open, decentralized and secure."

Building a Foundation of Trust

For an open standard to succeed, it needs a neutral, trusted steward. This is where the Linux Foundation's role is critical. "As the lead author of the award-winning Agent Name Service research paper, my greatest concern was that agentic AI systems would proliferate without a neutral, trustworthy identity and discovery layer," explained Ken Huang, CEO of DistributedApps.ai. "The Linux Foundation's governance model is precisely what ANS needs: vendor-neutral stewardship... that ensures no single actor can compromise the integrity of agent discovery at internet scale."

Furthermore, ANS isn't being built in a vacuum. The framework is designed to integrate with other emerging identity technologies, such as Decentralized Identifiers (DIDs) and Legal Entity Identifiers (LEIs). DIDs offer a way for agents to have portable, self-owned identities that aren't controlled by any single company, while LEIs provide a globally recognized way to tie an agent back to a verified corporate entity. This interoperability is key to creating a rich, multi-layered trust fabric for the entire digital economy.

The project is part of a broader, collaborative push to build the necessary safety and security layers for AI, including work from the Open Worldwide Application Security Project (OWASP) on agentic threats. It's a recognition that for this next phase of the internet to flourish, the groundwork must be laid now. As Srini Tallapragada, President at Salesforce, summarized, a common standard for identity and verification is imperative "so developers can move faster with trust, interoperability, and security built in from the start."