Zendesk has unveiled what it calls the Autonomous Service Workforce, a platform-level shift away from deflection-focused chatbots towards specialised AI agents that operate across every channel and are charged only when they produce a verified outcome.
The announcement, made at the company's annual Relate conference in Denver, frames the move as a direct response to what Zendesk describes as a “common industry failure”, in which organisations layer disconnected tools onto legacy workflows, optimising for ticket deflection rather than genuine resolution.
At the centre of the strategy is the Zendesk Resolution Platform, a unified system bringing together data, intelligence, knowledge, workflows, and governance. The platform has been trained on roughly 20 billion ticket interactions and operates through the “Resolution Learning Loop”, a mechanism designed to capture insight from every interaction in order to close knowledge gaps and improve automated responses in real time.
Beyond the chatbot era
Zendesk’s CEO, Tom Eggemeier, was direct about problem it is solving: "The era of the chatbot — the era of frustration and deflection — is over. We are entering the age of the Autonomous Service Workforce." Eggemeier described a future in which AI agents operate alongside human experts as a single, unified team, held to the same standards of accountability. The overarching ambition, he said, is to create “a future where AI is the foundation, and human experts are the architects."
Agent Builder, currently in early access, gives organisations a no-code interface to build, test, and deploy custom AI agents tuned to their own policies, workflows, and business logic. Zendesk's AI agents now operate across messaging, email, voice, and third-party AI platforms including ChatGPT and Gemini, maintaining shared context across interactions. Voice AI Agents, due in general availability later this quarter, support over 60 languages and can switch mid-conversation without losing continuity.
A new Employee Service offering, powered by Zendesk's recent acquisition of Unleash, extends the autonomous model to internal support. These agents work inside Slack and Microsoft Teams, search across enterprise systems, and enforce source-level permissions so employees only receive information they are authorised to access.
Copilots and continuous quality
Alongside the autonomous agents, Zendesk has expanded its Copilot portfolio for human teams. The Agent Copilot is designed to handle at least 30 per cent of tickets from day one by connecting to internal and external sources. An Admin Copilot, now generally available, helps administrators identify operational issues and apply workflow changes in real time. Knowledge and Analyst Copilots are in early access.
Quality Score, coming soon in early access, brings automated quality measurement to Suite Professional plans and above, analysing every human and AI interaction to surface performance gaps continuously rather than through periodic sampling.
Outcome-based pricing
The commercial model has been revised to match the platform's resolution focus. Zendesk charges only for interactions that are verifiably resolved, confirmed both by the AI agent and by an independent AI evaluation model. Spam and routine exchanges are excluded.
A pattern taking shape
The Zendesk announcement joins a series of major autonomous AI bets from enterprise software vendors. SAP set out its vision for an autonomous enterprise at Sapphire 2026 earlier this month, while Salesforce introduced Multi-Agent Orchestration to enhance shared context between channels and avoid customers having to repeat themselves. ServiceNow also recently launched Otto, a unified experience layer aimed at resolving fragmentation in AI deployments. The convergence suggests that the industry is moving away from isolated chatbots to harness AI in a way which will actually help to achieve resolution at scale.
Daniel Newman, CEO of Futurum Research, said the Zendesk approach recognises that "automation on its own is not enough". To improve customer experience in a meaningful way, he argues, “AI has to be part of a broader system that can connect context, take action, and evolve with the needs of the business”.