Freshworks expands Freddy AI with email & workflows

Freshworks has expanded Freddy AI in its service software, adding tools for workflow creation, email support and agent assistance.

Administrators can now create AI agent workflows in Freddy AI Agent Studio by describing a process in plain language. The system converts those instructions into workflow logic that can then be validated and added to an AI agent.

The change addresses a common obstacle in service automation: support teams may understand the steps needed to handle a request but still require specialist help to turn that process into a usable workflow.

Freshworks has also introduced Freddy AI Agent for Email, extending the product into a long-established support channel. The email agent is designed to manage multi-turn conversations, using context gathered across several replies until a resolution is reached.

The latest version of Freddy AI Agents also improves handling of files and images. In practice, that means the system can use information contained in attachments and pictures during the support process, which may reduce repeated exchanges between employees and service teams.

Another change affects how AI agents retrieve information. Filtered search on the design side gives administrators tighter control over the knowledge sources an agent can access, allowing teams to limit responses to material judged relevant for particular request types.

Copilot changes

Alongside the agent updates, Freshworks has shifted Freddy AI Copilot from a mainly on-demand assistant to a more continuous presence inside the service desk. Instead of waiting for a user to ask for help, the tool now surfaces information from knowledge bases and historical tickets as work progresses.

Freshworks has also added a conversational interface for Copilot. Support staff can ask follow-up questions, explore issues in more detail and use the system as a running guide during ticket handling.

Administrators have also been given additional controls over that experience. Teams can set clearer instructions for how Copilot should behave, what it should prioritise and how closely it should reflect internal service practices.

Broader updates

Beyond the headline changes, the update includes a Model Context Protocol gateway intended to make it easier for Freshservice to connect with external AI tools and third-party systems. That link is designed to give the software access to more live service context and actions.

Freshworks has also made a series of smaller product changes covering translation accuracy, ticket field suggestions, Writing Assistant behaviour and change recommendations within the Copilot experience.

Freddy AI Agent Studio has also been revised with custom functions, preset workflow triggers and richer workflow interactions. Those additions are intended to make it easier for teams to extend the behaviour of AI agents without rebuilding processes from scratch.

Service focus

The latest release shows how software suppliers are pushing AI deeper into day-to-day service operations rather than limiting it to narrow automation tasks or simple text generation. Here, the emphasis is on giving administrators more direct control over setup while widening the number of channels and situations in which AI can act.

For Freshworks, that means placing Freddy AI more firmly inside the flow of employee support. Workflow design, knowledge retrieval, ticket guidance and email exchanges are all being drawn into a broader model in which AI is present throughout a request rather than only at isolated steps.

Some of the newly discussed functions are being made available to select customers through the company's early access programme.

"What excites us most is the broader shift these updates represent. Freddy AI is becoming more useful across the service experience, more adaptable to the way each team operates, and more capable of helping move work forward with less friction. Each innovation step we take gives Freddy AI a more active role in service delivery and brings us closer to a future where AI can take on more work while teams stay in control and focused on better outcomes. That is the service operations future we plan on delivering."