AI agents are starting to talk to customers and to each other. There is no shared record of what an agent said, on whose behalf it spoke, or under what authority it acted. vCon is that record. It captures the dialog, the parties, and the consent in a single signed object that any downstream tool can verify.
Deepfakes blur the line between real and synthetic conversation. A spoofed call injected into an AI pipeline is the conversational analog of malware injected into a software supply chain. vCon pairs with SCITT, the IETF effort for Supply Chain Integrity, Transparency and Trust, so every lifecycle event (creation, sharing, analysis, deletion) is recorded in an append only ledger that cannot be altered after the fact.
Consent today lives in policies, contracts and screenshots. It rarely travels with the conversation it covers. vCon carries consent inside the file itself, scoped by purpose and time. That is the difference between reporting on a privacy policy and enforcing one. GDPR style deletion, AI training opt outs, and downstream sharing all become verifiable acts rather than promises.
Conversation data sits trapped inside contact center platforms, recording vendors, and AI tools that do not interoperate. The same vCon can be read by a contact center analytics tool, a compliance audit, and an AI agent without any of them agreeing on a private format first. The format is open, free of intellectual property encumbrances, and aimed at the next decade of conversation infrastructure.
A signed JSON object that carries parties, dialog (recording or transcript), attachments, analyses, consent, and a tamper evident history. The same format works for a phone call, a chat session, a video meeting, or a human to agent conversation.
The open source platform that creates vCons from business systems, runs them through pipelines of links (transcribe, redact, analyze, store, forward), and emits archive copies plus projections into the tools the business already uses.
Applications consume vCons through the Conserver API, the MCP server, or the language libraries. Adapters let CPaaS providers, contact centers, and recording vendors plug their platforms into the ecosystem without inventing a new format.