Zenphi launches Google Chat AI agents for enterprise workflows

Zenphi has launched AI Studio, a no-code builder for AI agents that let US-based Google Workspace users trigger governed business workflows directly in Google Chat. The preview release targets HR, finance, IT and compliance teams that want automation with audit trails and access controls, without the cost and complexity of traditional enterprise platforms. Why it matters: - Zenphi is aiming at a gap between basic workplace chatbots and expensive enterprise automation platforms. - The new AI Studio lets Google Workspace teams use Google Chat to start real business processes, not just ask questions. - The launch is designed for mid-sized organizations that need governed automation with auditability, permissions, and human oversight. What happened: - Zenphi announced Zenphi AI Studio, a no-code AI agent builder for Google Workspace users. - The product is now available in preview for US-based customers across all Zenphi subscription plans. - End users will interact with the agents through Google Chat. - The company positioned the launch as a way to turn chat into an action layer for enterprise workflows. The details: - AI Studio agents can submit leave requests, file reimbursement claims, start onboarding processes, report incidents, and assign tasks with automated reminders. - The agents can also answer questions using configured knowledge sources and company policies. - Agents can view, query, and respond to tasks, including approving or rejecting items waiting for action. - Existing Zenphi customers can add a conversational layer on top of workflows they already built. - New customers can build workflows and agents together in the same no-code environment. - Administrators can control which workflows an agent can trigger, which data sources it can access, which users or roles it can serve, and what information it can return under specific conditions. - Agent behavior can vary by user role and relationship to the underlying process or data. - An HR agent, for example, can show an employee their own requests and policy information while limiting a manager to approved information about direct reports. - A colleague outside that reporting relationship would not receive sensitive information. - Administrators get a central dashboard with usage, active sessions, conversation history, errors, workflow connections, and activity trails. - Zenphi said administrators can use that dashboard to investigate issues if a workflow connection breaks, usage nears a limit, or an agent acts unexpectedly. Between the lines: - Zenphi is pitching AI Studio as governed automation infrastructure, not a standalone autonomous assistant. - The company is leaning on compliance and control features to make the product usable in HR, finance, legal, IT administration, and other sensitive functions. - Zenphi is also trying to differentiate from large enterprise service platforms that can handle similar needs but often require substantial implementation budgets and infrastructure. - Vahid Taslimi, Zenphi’s CEO, said the platform is built for agents that execute with governance and audit trails, not just assist. - Taslimi also said Zenphi’s platform already runs over 1.5 million AI-powered tasks a month for organizations including Google, Gordon Food Service, Action Behavior Centers, and North Carolina State University. What’s next: - Zenphi AI Studio is in preview, so broader rollout details have not been announced. - The company said governance rules are configured separately for each new use case as agents expand. - Zenphi will likely use the preview period to prove that Chat-based agents can safely handle real operational work inside Google Workspace environments. The bottom line: - Zenphi wants Google Chat to become a controlled execution layer for enterprise work, with AI agents that can take action inside governed workflows instead of just generating answers. - The platform’s compliance certifications, Google Cloud Partner status, and existing production usage are central to that pitch.