"The road to becoming an AI creator | Article introduction" What is an AI agent that evolves customer experience? A thorough explanation of the mechanism, use cases, and points to note! Full of tips for improving CX. #AIAgent #CustomerExperience #GenerativeAI
Video explanation
How will AI agents and generative AI change the future of customer experience (CX)? A comprehensive guide for beginners
Hello, I'm John, a veteran blog writer. Today, I'd like to explain some of the more commonly heard terms these days: "AI agents," "customer experience (CX)," and "generative AI." These terms may seem a little difficult to understand, but I'd like to explain them in a way that's easy for beginners to understand. Let's take a look at how these technologies could potentially change our everyday shopping and service experiences!
Introduction: AI Agents and Customer Experience (CX), What is Generative AI?
First, let's look at the basic meanings of words.
- AI agentsSimply put, an AI is an intelligent software program that performs certain tasks automatically, either for us or to help us, such as managing our schedules, gathering information, or operating complex systems.
- Customer Experience (CX)Customer experience refers to the entire series of experiences that a customer has from coming into contact with a company's product or service, purchasing it, using it, and receiving support afterwards. If this experience is good, the customer will be satisfied and will want to use the company's product or service again.
- Generative AI: A type of AI that can "generate" original content, such as new text, images, music, and even program code. For example, it can respond to questions in natural language and create new design proposals based on instructions. ChatGPT, which has been a hot topic recently, is one such example.
The combination of these technologies, particularly AI agents and generative AI, promises to enable businesses to provide smoother, higher-quality customer experiences that are tailored to each individual.
Why AI agents are gaining attention: SaaS platforms and improving employee experience
While AI once garnered attention for its large language models (LLMs) on the consumer front, things are moving a little differently in the enterprise world, where AI agents are increasingly being used to improve employee experience and productivity, especially through SaaS (Software as a Service) platforms.
For example, AI agents are beginning to help human resources departments streamline recruitment activities, marketing departments individually optimize advertising campaigns, and IT service desks respond to help desk inquiries, allowing employees to focus on more creative and important tasks.
The Potential for AI Agents to Revolutionize Customer Experience (CX)
AI agents have been making progress in improving employee experience, and it is said that it is only a matter of time before they become the standard in the field of customer experience (CX) as the next frontier. AI agents may dramatically improve the hassle of confusing operation screens, complex search tools, and long data entry forms. The future where AI agents understand customer preferences and situations and provide proactive support is approaching.
"Businesses that want to add value from AI to their customer experience will deploy specialized AI agents with expertise in specific areas such as product lines, inventory, pricing, shipping, and legal constraints," said John Kim, CEO of Sendbird. "We're already seeing this shift in industries such as retail, where AI is improving the shopping experience through personalization and proactive service. In the future, consumers will likely have their own personal AI assistants or multiple agents for finance, entertainment, healthcare, travel, and more."
First step: Start by automating repetitive, tedious tasks
In the past, there have been cases where companies rushed to provide AI-based customer experiences, which ended up negatively impacting customers and brand image. For this reason, many companies are cautious about establishing prerequisites such as AI governance (guidelines and systems for dealing with ethical, legal, and social issues related to AI), data quality, and thorough testing before using AI agents to improve customer experience.
Early adoption opportunities include:Large-scale, repetitive, "boring" tasks that are limited in scope and tend to frustrate customersIt would be a good idea to pay attention to the following.
"The biggest impact of generative and agent-based AI is automating the most tedious and repetitive CX micro-workflows," notes Dave Singer, global VP at Verint. "A variety of CX tasks, like asking the right questions to get relevant information at the start of a customer interaction, finding answers to customer questions, and post-call wrap-up, can be automated with specialized AI-powered bots to drive stronger and faster business outcomes. The result is freeing up human agents, improving the CX, and reducing costs, increasing revenue, or both."
For example, think about the manuals and tools your customers use to learn about your product, install it, or solve problems: instead of having your customers sift through pages of documentation, it can be a much faster and easier experience to have an AI agent answer their questions.
"Think about the 'finding pads' your customers visit along the way, like product help pages, user wikis, and online communities, and consider how generative AI or LLMs can improve them to enhance the customer experience," suggests John Kennedy, CTO at Quickbase. "Make your solutions easier to use and more effective with a library of pre-built templates that can be accessed with a few prompts, all to serve customers by industry, role, or other segment. Think about your customers' continuous learning and how AI can help you leverage the experiences built in your customer communities to guide them to the next milestone and improve their journey."
Deon Nicholas, founder and president of Forethought, says you should find easy ways to handle simple customer tasks, rather than just presenting information quickly: "One of the easiest user experiences to develop with LLM is a chatbot that can provide RAG-based search and quickly surface information from FAQs in response to customer questions. But you can achieve even better ROI by incorporating agent-like AI into your web or app experience to take actions on your customers' behalf, like resetting a password or checking the status of an order."
The key to utilizing AI: centralizing and quality control of customer data
To use AI for more interactive customer experiences, companies need to use centralized, cleansed data to train and test their AI agents. Companies are using tools like customer data platforms (CDPs) and data fabrics to connect customer data and interactions.
"A robust AI-driven CX strategy is only as good as the underlying data and associated governance. Any CX program should prioritize a continuous testing and learning strategy to ensure data freshness and accuracy," said Tara Desao, senior director at Pega. "This approach not only improves agent performance, but also reduces risk and increases brand value as consumers interact with companies across channels."
When centralizing customer data, data controls must be established around security, user access levels, and identity management. Many organizations turn to Data Security Posture Management (DSPM) platforms to help mitigate the risks of managing numerous data sources, multiple cloud database platforms, and disparate infrastructures.
"By rethinking how data is stored and accessed, and moving from siloed third-party systems to a user-centric data model, organizations can create more fluid, responsive web and mobile interactions that adapt to preferences in real time," said Osmar Olivo, VP at Inrupt. "To maintain accuracy and performance, AI-driven experiences should be trained on diverse real-world data while incorporating user feedback mechanisms that allow individuals to modify, refine and guide AI-generated insights by providing their own preferences and metadata."
Manish Rai, VP at SnapLogic, predicts that over 80% of generative AI projects fail due to issues with data connectivity, quality, or reliability. "Success will depend on tools that simplify agent development, AI-enable data, and ensure reliability through observability (external visibility into the internal state of the system), accuracy assessment, and policy enforcement."
Rosalia Siripo, VP of Data Science Evangelism at KNIME, points out that many agent-based applications have a human-in-the-loop step to check correctness: "In other cases, special guardian AI agents focus on controlling the results, pushing them back and asking for improvements if they are not satisfactory." For data-related tasks such as sentiment analysis, "the accuracy of the generative AI is compared to that of other classical machine learning models."
Evolving customer service: Improving phone and chat response with AI agents
Beyond finding information and performing simple tasks, customer service calls and chats are a pain for both consumers and the human agents they interact with. In one survey, 23% of respondents said they would rather watch paint dry than go through repeated bad customer service experiences.
Instead of rule-based chatbots with limited capabilities, customer service AI agents can sift through the data and respond to customers, while human agents can take on the more difficult cases with the help of customer success AI agents.
"There's a clear correlation between customer satisfaction and effective self-service usage," said Vinod Muthukrishnan, VP and COO, Webex Customer Experience Solutions, Cisco. "The evolution to true agent-like AI transforms the self-service experience by orchestrating the end-to-end engagement between your brand and your customer. This advanced AI capability empowers customer experience teams to deliver intelligent, seamless interactions and meet customers where they are, on their schedule."
Part of the challenge is data, and another is that many customer experiences were developed as point solutions that only addressed part of the customer journey. As technologists move to generative AI-enabled engagement, they should apply a design thinking approach to redesign more holistic experiences.
"Customer-facing applications like websites, mobile apps, and B2C messaging typically have back-end integrations with customer-specific data sources that enable the applications to answer questions and solve problems," said Chris Arnold, VP of Customer Experience Strategy at ASAPP. "Leveraging LLM to curate a conversational, personalized experience far surpasses the transactional experiences these applications can provide on their own."
Essential before deployment: Thorough testing of AI agents
Organizations looking to develop more advanced CX capabilities or autonomous AI agents will need a comprehensive testing plan to validate their functionality. Prompt filters, AI response moderation, content safeguards, and other guardrails can help CX agents avoid inappropriate or out-of-bounds conversations. But brands need to go beyond these basics to ensure their CX AI agents respond appropriately, accurately, and ethically.
"We can't just ship agents out into the world without testing and monitoring," stresses Sada CTO Miles Ward. "Rigorous testing for accuracy and performance is non-negotiable. We need to make sure they provide a smooth and reliable experience. Otherwise, we're just creating new problems."
Ganesh Sankaralingam, data science and business analytics lead at LatentView, says AI experiences and LLM responses should be tested for accuracy and performance across five dimensions:
- Relevance: Measures how appropriate and relevant a response is to the question.
- Groundedness: Evaluates whether the response is consistent with the input data.
- Similarity: Quantifying how close the AI-generated response is to the expected output.
- Coherence: Evaluate the response flow to ensure it mimics human-like language.
- Fluency: Assess the language proficiency of the response to ensure that it is grammatically correct and uses appropriate vocabulary.
"Businesses should test the accuracy and performance of their AI experiences by running their AI agents against past customer questions and reviewing the results," says Deon Nicholas of Forethought. "It's also important to measure how often the AI can handle customer interactions autonomously, and apply separate evaluation models to review the sentiment and accuracy of the conversations."
AI Agents and the Future of Customer Experience
How will AI agents impact customer experience (CX) in the near future? Mo Sherif, Senior Director of Generative AI at Sitecore, recommends rethinking the entire experience: "To create a truly agent-like experience, don't just augment what you already have, but build your journey specifically as a generative AI-first interaction."
There are contrasting views on how AI agents will evolve. Some predict a more autonomous future where people empower and trust AI agents to make more complex decisions and take a wider range of actions. Others predict a more human-centric approach where AI agents augment human capabilities and partner with people to make smarter, faster and safer decisions.
Michael Wallace, Americas solutions architecture leader, customer experience, Amazon Web Services (AWS), says agent-based AI can resolve issues without human intervention. Think of a contact center that self-heals during a crisis, automatically reallocating resources, updating customer communications, and resolving issues before customers experience them.
"Imagine an airline facing a sudden surge in traffic due to weather delays," said Wallace. "With agent-based AI, the contact centre could make autonomous decisions about rebooking or proactively notifying passengers, freeing up human agents to focus on complex customer needs rather than administrative tasks."
Doug Gilbert, CIO and chief digital officer at Sutherland Global, says AI shouldn't be about automating customer experiences, but making them more human and intelligent: "The true value of generative AI isn't in replacing human interactions, but in augmenting them to make them smarter, faster and more natural. The trick is AI that learns from real-world interactions and constantly evolves to feel less robotic and more intuitive."
Both autonomous and human-facing CX AI agents are likely a reality, but in the meantime, companies need to thoroughly research customer needs, improve data quality, and establish rigorous testing practices.
Companies and experts driving AI technology
As mentioned in this article, the fields of AI agents, customer experience, and generative AI are being driven by the efforts of many leading companies and experts, including Sendbird, Verint, Quickbase, Forethought, Pega, Inrupt, SnapLogic, KNIME, Cisco, ASAPP, Sada, LatentView, Sitecore, AWS, and Sutherland Global, all of whom are contributing to the application and development of AI technology in their respective fields of expertise.
The experts at these companies are exploring the possibilities of AI and providing knowledge and solutions to realize better customer experiences. By paying attention to their comments and research and development trends, you can learn about the cutting edge of AI technology. This is a very active field, and we expect to see many innovations in the future.
Frequently Asked Questions (FAQ)
- Q1: What exactly does an AI agent do?
- A1: An AI agent is software that acts autonomously to achieve a specific goal. For example, it can perform a wide range of tasks, such as automatically answering customer inquiries, recommending products, handling reservations on behalf of customers, and providing initial support for complex problems. This reduces the burden on humans and enables faster, more personalized service.
- Q2: Are generative AI and AI agents the same thing?
- A2: No, it is different. Generative AI is an AI technology that creates new content (text, images, etc.). On the other hand, AI agents are the entities that perform tasks and may use generative AI within them. For example, an AI agent may utilize the text generation capabilities of generative AI to have natural conversations with customers.
- Q3: What are the main benefits of introducing AI agents into your customer experience?
- A3: The main benefits are:Available 24 hours a day, 365 days a year,Improved response speed,Providing a personalized experience,Cost reductionAndEmployee load reductionThis is expected to improve customer satisfaction and business efficiency.
- Q4: What should companies pay attention to when introducing AI agents?
- A4: The most important thing is,Ensuring data qualityとEstablishing an appropriate governance systemAI learns based on data, so inaccurate data may produce erroneous results. In addition, it is necessary to pay careful attention to ethical issues and privacy protection, and to operate the system while maintaining transparency.Extensive testingis also essential.
- Q5: Is it possible for small and medium-sized enterprises to introduce AI agents?
- A5: Yes, it is possible. In recent years, there has been an increase in cloud-based AI agent solutions that are relatively affordable. You can start by automating small tasks, and consider introducing them in a way that suits your company's size and purpose. The important thing is not to aim for a large system right away, but to start small and verify the effectiveness as you go.
Summary and future precautions
AI agents and generative AI have great potential to fundamentally transform customer experiences. A future where more personal, more efficient, and more satisfying service is the norm may not be far off. But to maximize the benefits, businesses need careful planning, quality data, and rigorous testing.
In particular, there is a non-zero risk that AI will provide incorrect information or respond in an unexpected way. For this reason, it is important to start with a limited scope in the early stages of implementation, operate under human supervision, and make continuous improvements.
I hope this article helps you better understand AI agents, generative AI, and the future of customer experience they bring.
Disclaimer:This article is intended to provide information and is not intended to recommend the adoption of any particular solution. When considering the adoption of AI technology, we recommend that you conduct your own thorough research and comparative studies, and also consult with experts.
Related information/For those who want to know more
- Infoworld: About Large Scale Language Models (LLM)
- Infoworld: About AI Agent Development
- Official websites and technical blogs of each company (Sendbird, Verint, AWS, etc.)
- Specialized media and research reports on AI and customer experience