GoDaddy Joins a Second AI Agent Identity Consortium. DNS Is the Foundation.

When an AI agent visits a website today, the site owner typically cannot tell who sent it, what it is authorized to do, or whether it is legitimate. Training crawlers from OpenAI and Anthropic are consuming web content at a crawl-to-referral ratio of thousands to one, sending almost no traffic back to the publishers whose content they consume. Cloudflare reports that its customers are already sending over one billion HTTP 402 “payment required” responses per day to AI crawlers, a signal that the market is trying to price access long before a billing mechanism exists. The infrastructure companies that process this traffic are moving fast to establish the rules. On May 7, 2026, GoDaddy became a founding partner of the HOL (Hashgraph Online) Partner Program, joining a consortium of more than 30 companies building open standards for AI agent identity on DNS. The announcement came exactly one month after GoDaddy and Cloudflare announced a separate but related partnership on April 7. GoDaddy is now at the center of both.

The Problem: No One Knows Who the Agent Is or Who Sent It

AI agents interact with websites the same way browsers do. They make requests, retrieve content, and move on. Unlike a human visitor, an agent may be placing an order, extracting training data, or probing for vulnerabilities. A website owner has no reliable way to distinguish between them. The agent has no verifiable identity, no business affiliation that can be checked, and no accountability if it acts outside its intended scope.

GoDaddy’s response is the Agent Name Service (ANS), an open standard it has been running in production since November 2025. ANS assigns each AI agent a unique, human-readable name and a verifiable identity tied to a domain, using the same infrastructure that already powers the web: DNS and digital certificates. The principle mirrors how websites already establish identity: through domain ownership and a certificate that proves the domain is genuine. ANS applies the same logic to agents. LegalZoom registered the first agent on ANS on April 2, 2026, a legal services integration that allows AI assistants to connect users with attorneys through a verified, named identity. LegalZoom Chief Customer & Business Officer Aaron Stibel described the rationale: “AI agents will transform how legal services are delivered, but their value depends on verifiable identity and human accountability.”

Two Partnerships, One Infrastructure: What GoDaddy and Cloudflare Built in April

The April 7 partnership between GoDaddy and Cloudflare has two components. The first is AI Crawl Control, a Cloudflare tool being integrated into GoDaddy’s hosting platform. It gives website owners three choices for each AI crawler: allow access, block access, or display an HTTP 402 response indicating that access requires payment. The tool became generally available in 2025 and now handles over one billion such signals daily across Cloudflare’s network.

The second component is agent identity. Cloudflare’s Web Bot Auth standard uses cryptographic signatures in HTTP requests to verify that a request comes from the agent it claims to be. GoDaddy’s ANS provides the directory where agents are registered and named. Together, they create a mechanism where a verified, named agent can arrive at a website, identify itself, and be granted or denied access based on policy. GoDaddy Chief Strategy Officer Jared Sine described ANS as “an open standard giving every agent a verifiable identity built on DNS.”

What the HOL Partnership Adds: A 30-Company Consortium and a Second Trust Layer

The May 7 HOL announcement adds a different dimension. HOL (Hashgraph Online) is an open-source platform building coordination standards for AI agent infrastructure. Its founding partner cohort includes more than 30 companies operating across five areas: agent registries and identity, payments and commerce, privacy and security, inter-agent communication, and tooling. GoDaddy participates as a founding partner in the registries and identity category. HOL’s registry currently indexes 249,000 agents and 24,000 MCP servers and has published more than 20 open standards.

GoDaddy and HOL jointly published two draft specifications. The first, HCS-14, defines a universal identifier format for AI agents that works across both traditional web services and decentralized systems, so that an agent registered in GoDaddy’s ANS can be recognized and verified by systems outside the traditional web. The second, HCS-27, creates a publicly verifiable audit trail for the agent registry: at regular intervals, a cryptographic snapshot of the registry is published to a shared ledger, allowing any party to confirm that the registry records have not been altered without trusting any single company’s word. The ledger used is Hedera’s consensus network, which HOL uses across its protocol stack.

GoDaddy VP of Engineering Scott Courtney described the practical effect: “These drafts let any resolver discover and verify ANS-registered agents through a standard interface, and let anyone audit the registry’s history without trusting a single operator.” HOL President Michael Kantor framed the goal as interoperability: the Universal Agent ID creates “a shared discovery layer that makes agents easier to find and connect across environments.”

The two partnerships are complementary rather than competing. ANS is the identity system in both. The Cloudflare integration makes ANS operational in traditional hosting environments. The HOL standards extend ANS into systems that operate outside the traditional web, including decentralized platforms that use shared ledgers for trust. A business running agents that interact with both traditional web services and decentralized ecosystems benefits from an identity standard that works in both.

The Hosting Industry Is Already Splitting on AI Access. Standards Would End the Guesswork.

Hosting providers are already making AI crawler decisions for their customers, often without telling them. A May 2026 investigation by Search Engine Land documented how several major managed WordPress hosts handle AI training crawlers differently, with results that surprised many site owners.

WP Engine blocks AI training crawlers including ClaudeBot and GPTBot at the platform level, returning HTTP 429 responses. The blocks fire below the layer where customers can see them, meaning standard auditing tools show no issue. Customers cannot disable the blocks per-bot through the control panel, and the company excludes suspected bot traffic from billable metrics without notifying customers of the policy.

Kinsta took the opposite position. Its CTO stated in March 2026 that the company will not block at the platform level and will not charge for bot bandwidth. Kinsta offers an opt-in Bot Protection feature with four customer-controlled levels. Pressable and Pantheon also do not block by default, leaving crawler management to the customer. SiteGround blocks AI training crawlers by default but is more transparent about the policy than WP Engine.

The split illustrates the absence of a common standard. Without a way to verify agent identity, hosting providers face a binary choice: block all unverified agent traffic (and risk blocking legitimate agents) or allow all of it (and absorb the cost and content risk). ANS and the HOL standards are designed to eliminate that binary by making agent identity checkable before a decision is made. A hosting provider that can verify an agent’s identity and its operator can offer a policy engine rather than a platform-level blanket decision.

GoDaddy has positioned itself to be that policy engine for its 20.4 million customers. Its planned ANS service offering covers agent hosting, registration, certificate management, a discovery marketplace, and analytics, layering a new revenue stream on top of its existing domain and hosting business. Every AI agent that wants verifiable access to a website needs a registered identity. If that identity is anchored to a domain, and the domain is managed through a registrar, the registrar becomes a required step in every agent’s commercial relationship with the web.

The standards are still drafts. The two consortia are still being built. But the direction is clear: the companies that established how humans navigate the web are now competing to define how agents do. The companies with the most structural advantage in that race are the ones that already own the infrastructure those standards are being built on.