Vint Cerf's Next Mission: Giving AI Agents a Verifiable Identity on the Open Internet

What happened

Vint Cerf — one of the architects of the TCP/IP protocols that underpin the open internet — retired from Google on July 7, 2026, after roughly 21 years as the company's chief internet evangelist. But he's not stepping away from the internet's future. Starting July 15, Cerf is advising Innovation Labs, an organization working to build open architecture for AI agents to identify themselves on the public internet.

Innovation Labs is a subsidiary of Identity Digital, a DNS registry company. Its proposed standard, called DNSid, is a registry for agent identification that links each AI agent to an existing internet domain name and uses cryptographic proofs to verify its identity. The idea is to use the domain-name infrastructure that already runs the web as a practical mechanism for holding AI agents accountable.

Cerf joins a handful of other internet luminaries lending their names to the effort. The timing aligns with a structural shift he has been warning about: the rise of autonomous AI agents will require the same kind of formal, open interoperability standards he helped build in 1974.

Why it matters

Most AI agents today operate inside proprietary systems — think of an agent calling internal APIs within a single company's platform. But businesses are already envisioning a world where agents operate far more autonomously, crossing organizational boundaries and interacting directly with other agents across the open internet. The key roadblock so far has been the absence of a shared standard for identifying and auditing those agents.

Without such a standard, there's no reliable way to know whether an agent knocking on your digital door is who it claims to be, who it works for, or what it's authorized to do. DNSid's bet is that the domain-name system — already globally deployed and trusted — can serve as the identity layer for this new agent-to-agent economy.

The stakes extend beyond identity. The broader context is an internet where AI is reshaping the economics of the open web. By 2026, roughly two-thirds of all US Google searches end without a user visiting a third-party website, according to SparkToro and Similarweb's zero-click search study. For news-related queries, that figure rose from 56 percent to 69 percent in the year following Google's launch of AI Overviews. If agents become the primary actors online, the infrastructure that governs them will determine whether the next era of the internet remains open — or fragments into walled gardens.

What to watch

• DNSid adoption trajectory. Innovation Labs has proposed the standard, but it's early. Watch for whether major AI platforms, cloud providers, or enterprise vendors signal support or propose competing frameworks.
• Competing standards landscape. A variety of agent-identification standards are beginning to emerge. DNSid is one proposal among several; convergence or fragmentation will shape how quickly cross-platform agent interoperability becomes real.
• Cerf's influence. His involvement lends credibility, but the question is whether that translates into industry-wide adoption or remains an advisory role.
• Regulatory attention. As agents increasingly act on behalf of people and businesses, expect policymakers to take interest in accountability frameworks — which could either accelerate or complicate voluntary standards like DNSid.