- Products›
- Amazon Connect›
- Amazon Connect AI Agents
Amazon Connect AI agents
Transform customer service with agentic AI
Meet the future of customer experience: AI agents that understand, reason, and take action
AI agents dynamically navigate your organization’s resources to find solutions and take action to resolve customer needs. They handle many issues on their own, but also work in collaboration with your workforce to deliver personal, effortless customer experiences.
Help your customers get answers and resolve issues instantly
Customers have natural conversations with AI agents that already know their history and preferences via their preferred channel. Instead of just providing information, AI agents can actually take action, handling everything from returns to account updates. This means issues are resolved quickly, without transferring customers between systems.
Empower your workforce to deliver exceptional service
AI agents tackle time-consuming tasks, like summarizing email threads or case histories, that traditionally slow down your teams. The result? Your workforce is able to focus on what really matters: delivering incredible customer experiences.
Configure and deploy AI agents with confidence
Deploy first-party and custom AI agents at scale using the Amazon Connect AI agent designer. Using Model Context Protocol (MCP) support, choose what information your AI agents can access, from customer profiles and case histories to your knowledge base and third-party business systems, so they take action within your established guardrails.
Monitor AI agent performance
Track every decision, understand every action, and see exactly how your AI agents deliver results. View performance in intuitive dashboards to optimize performance, ensure compliance, and build trust in your AI-powered experiences.
Customers
Anthology
Anthology is an education technology company that provides solutions for higher education institutions.
"[…] The ability for the Connect assistant to leverage external tools through Model Context Protocol (MCP) will make it truly agentic, giving us the potential to empower students and staff with personalized recommendations and the ability to take immediate action without leaving the chat interface…”
Richa Batra, Senior Vice President and General Manager of Anthology Student Success.
Zepz
Zepz empowers customers with seamless cross- border payments, connecting over 100 million people across more than 130 countries.
“With chat being a primary channel, the AI- powered Connect assistant delivers quick, hassle- free resolutions, successfully addressing 67% of customer inquiries instantly through self-service. By using Amazon Connect Cases and Customer Profiles, our agents—both AI and human—can easily access a customer’s full transaction history, ensuring every interaction is highly personalized and customers never have to repeat themselves."
Mark Russell, Contact Centre Architect, Zepz
Orbit
Orbit Irrigation is a manufacturer and supplier of home and commercial irrigation systems.
"To resolve customers' questions, our agents spend two to three minutes per interaction searching through several different sources of knowledge, including Orbit product pages, customer account pages, and internal knowledge forums. This multistep process adds time to the interactions for agents and customers. The new responses automatically generated at each turn of the customer conversation by the Connect assistant are tailored based on our own knowledge base articles. The Connect assistant will create 10%–15% time savings on every contact, and the increased number of calls handled every hour is expected to translate directly into costs savings for Orbit—all done with improved customer sentiment."
Brian Dick, Senior Manager of Customer Care, Orbit Irrigation
Frontdoor
Frontdoor, a leading provider of home warranties and digital on-demand services, is piloting the Connect assistant with the intent of reducing agent training and on-boarding time. This pilot is already reducing an agent’s path to proficiency by delivering agents’ next best responses and actions, based on policy documents stored in Amazon S3 within the Amazon Connect agent workspace. As they learn from how agents use this technology, Frontdoor expects to expand this same support to consumer-facing self-service interactions.
Pronetx
Pronetx, a professional services partner for customer experience transformation, is implementing Amazon Q in Connect for a number of public sector, federal, and FinTech organizations. With Amazon Q in Connect, they have the ability to use a single engine to drive both customer facing conversational self-service experiences and context-aware suggestions and automations for representatives. Pronetx expects Amazon Q in Connect will allow the public sector, federal, and FinTech organizations they support to focus on creating the experiences that matter most to their customers while empowering their representatives with the best decision-making information and guidance at the right time in every customer touchpoint.
FAQs
How can I learn more about Amazon Connect?
For more information, see Amazon Connect.How much does it cost to use Amazon Connect AI agents?
To review the pricing information, see Amazon Connect Pricing.
What happened to Amazon Q in Connect?
Since we first launched Amazon Q in Connect, AWS’ generative AI assistant for customer service, AI technology has evolved rapidly. Amazon Connect now enables you to create and configure multiple AI agents, including out-of-the-box and third-party agents. To make it easier to understand these expanded capabilities, functionality previously known as Amazon Q in Connect is simply one of our AI agents.
What is the relationship between the Connect assistant and AI agents?
Customers and human agents interact with one or more AI agents through a conversational interface called the “Connect assistant.” The Connect assistant is embedded in the Amazon Connect agent workspace for your human agents, and surfaced externally for your customers.
How does Model Context Protocol (MCP) work within Amazon Connect?
Amazon Connect supports Model Context Protocol (MCP), enabling AI agents to use standardized tools for retrieving information and completing actions. Out-of-the-box MCP tools can be used for common tasks such as updating contact attributes and retrieving case information. Customers can also create new or convert their existing flow modules into MCP tools, enabling them to reuse the same business logic across both static and generative AI workflows.
Do AI agents work with Amazon Lex?
Yes, AI agents works seamlessly with Amazon Lex. While AI agents can handle more complex, open- ended interactions, Amazon Lex remains a foundational service for handling predefined intents and specific conversational flows. The two capabilities complement each other, allowing businesses to leverage the strengths of both traditional conversational AI and advanced agentic AI within their self- service experiences.