Autonomous Agents are a GO

Jan 2, 20255 mins read

The future of service delivery just got a lot more intelligent

We saw something at the recent Microsoft Ignite conference that really blew our minds and opened our eyes: autonomous agents are now being put to use in Copilot Studio and will soon be available in Microsoft Dynamics 365.

What are Microsoft autonomous agents?

Simply put, autonomous agents are AI-powered systems that operate independently to handle a wide range of tasks. From real-time translations to project management to qualifying prospects, the potential applications are nearly limitless. Let’s explore the specific benefits of autonomous agents currently available in Copilot Studio, and those anticipated in Dynamics.

Autonomous Agents in Copilot Studio

Copilot Studio is the ideal launching pad for Microsoft autonomous agents, having been derived from the company’s low-code solution for intelligent chatbots. But what’s happening now with autonomous agents feels like light years beyond.

It is all made possible by leveraging Copilot Studio capabilities, including:

  • Asynchronous orchestration of complex tasks: Microsoft has introduced Copilot Actions, enabling users to automate repetitive tasks using natural language prompts. This feature allows for the automation of complex, multi-step processes, enhancing efficiency.
  • Memory and context: Copilot agents can access and utilize organizational knowledge and data sources, ensuring interactions are informed by relevant context. This Dynamic agent capability reduces the need for users to repeatedly provide information, streamlining workflows.
  • Monitoring, learning and improving: Copilot agents are designed to learn from interactions, applying insights in real-time to enhance their effectiveness and the quality of service delivered. The Copilot Studio platform includes features that allow for monitoring agent performance and making iterative improvements.
  • Delegation with confidence and guardrails: Autonomous agents operate within user-defined instructions and adhere to organizational security policies. The Copilot admin center provides tools to manage and monitor these agents, ensuring they act within established parameters.

Users of Copilot Studio can use new autonomous agents that can extend CRM solutions (such as Dynamics 365 Sales, Service and Contact Center), acting in the background to respond to events, like receiving an email or remembering a previous related conversation, without the need for wasted human time and effort.

Creating these agents is made simple in Copilot Studio as Microsoft provides a templated approach based on commonly used agent scenarios, such as taking action when a specific event occurs, for instance when an email is received or a new employee being added to the HR system. Let’s use email as an example.

An email arrives and the autonomous agent kicks into action, understanding the intent of the email, and/or looking up the sender's account, and/or checking inventory, and/or processing an order, and then taking the appropriate actions to close a ticket, orchestrating and shepherding the entire process over days.

Remember when it came to solving everyday business challenges how some would say, “there’s an app for that?” Now, it’s even better: with Copilot Studio we will soon get to a point where we say, “there’s an autonomous agent for that,” and whatever “that” may be will have been handled automatically, in real-time, and with limited (if any) manual intervention.

Autonomous Agents in Dynamics

The initial focus for the use of autonomous agents in Microsoft Dynamics 365 will be in customer service, where AI is unlocking opportunity after opportunity. When Copilot was first being put to use in the customer experience, it served primarily to boost agent productivity by, for example, making it faster and easier for agents to access and share critical customer information. No more lengthy database lookups: just a natural language request to Copilot.

Microsoft has introduced a range of pre-built agents designed to enhance productivity across various business functions. These agents can be deployed immediately or further configured to incorporate specific organizational knowledge and business logic. But with autonomous agents in Dynamics, customer service is significantly streamlined, evidenced by Customer Intent Agent, Customer Knowledge Management Agent, and Case Management Agent.

  • Customer Intent Agent: According to Microsoft, this agent will engage with your customers in a channel of their choice and, through natural language, determine the intent of a customer’s inquiry. It can reference all prior conversations, use all available knowledge to determine the right intent. With this agent, you will know what a customer requires without even speaking to them first. It references prior conversations and available knowledge to accurately assess customer needs.
  • Customer Knowledge Management Agent: Keeping a knowledge base up to date is one of the toughest jobs in the business. But imagine a knowledge base that can continually update its answers and articles to frequently asked questions (the FAQs) to address any customer question. The Customer Knowledge Management Agent can deliver exactly this, perpetually updating existing knowledge base content by analyzing case notes, transcripts, summaries, and other info from customer cases, and even creating entirely new articles and answers for new customer queries.
  • Case Management Agent: Working in the background during a customer interaction, this autonomous agent extracts vital information, fills out all forms, creates case notes, and more. While a human being can review, approve, and edit the case create, the majority of the task is handled by the autonomous agent.

The Case Management Agent isn’t merely a note taker, though. It can create cases, entering product/service information, creating and updating a case status, building communications, and more. Once updated, a human can still review, approve and edit the case that was created.  Since most of the work is initiated by the autonomous agent, the result is huge productivity gains for human agents as they no longer have to “do the paperwork. An autonomous agent does the heavy lifting, reducing the effort and time required by the human.

What do autonomous agents mean for service intelligence?

As you may have gleaned, the most obvious and immediate benefit realized in service intelligence will be automation collection and analysis of service-centric data. Autonomous agents can sift through massive amounts of data from various sources (customer interactions, system logs, social media, etc.) much faster and more efficiently than humans.

Furthermore, they can detect subtle patterns and trends that might be missed by human analysts, and predict future trends, such as potential service disruptions, customer churn, or emerging issues, allowing for proactive intervention.

Combine those improvements with increases in efficiency, cost-savings, and quality of service, and a revolution in service intelligence now appears to be in our immediate future.