"DNS forms the technical foundation. Both companies are calling on cloud and platform providers to join the joint standardization effort within the IETF. AI agents are increasingly operating autonomously on websites, in enterprise applications, and in cloud environments. But how does a system know which agent it is communicating with and whether that agent is trustworthy? That is precisely the problem Infoblox and GoDaddy aim to solve with two complementary open standards: DNS for AI Discovery (DNS-AID) and Agent Name Service (ANS)."
"DNS-AID, currently under review as an IETF draft, describes how AI agents can publish discoverable metadata via existing DNS record types such as SVCB, DNSSEC, and DANE. The standard is not controlled by any single party and can be implemented by any organization, platform, or agent framework. ANS focuses on the agent's identity. GoDaddy is a co-author of the ANS IETF draft. The standard enables agent owners to use existing domain names and provides cryptographic verification via PKI."
"Infoblox and GoDaddy emphasize that trust decisions must be based on open, cryptographically verifiable signals, not on proprietary reputation scores from a single vendor. Agent owners must retain control over identity, metadata, and policy. Both Infoblox and GoDaddy invite cloud providers, agent platforms, registrars, and standards organizations to participate in IETF work."
""Adopters of the Agent Name Service open standard leverages the only infrastructure that exists today that operates at the scale and speed of the global internet - Domain Name Service," said Jared Sine, chief strategy and legal officer at GoDaddy."
AI agents need a way to identify themselves, be located, and prove trustworthiness when communicating on the open web. DNS provides the technical foundation for two complementary open standards. DNS for AI Discovery (DNS-AID) enables agents to publish discoverable metadata using existing DNS record types such as SVCB, DNSSEC, and DANE. Agent Name Service (ANS) focuses on agent identity by letting agent owners use existing domain names and verify identity through PKI-based cryptographic mechanisms. The standards are designed to be open and implementable without a single controlling party or closed registry. Trust decisions should rely on open, cryptographically verifiable signals rather than proprietary reputation scores, while agent owners retain control over identity, metadata, and policy.
Read at Techzine Global
Unable to calculate read time
Collection
[
|
...
]