With artificial intelligence (AI) becoming increasingly widespread, many business owners likely feel that it has helped reduce their operational workload. AI is an incredibly useful tool, and it’s expected to have a significant impact on our decision-making processes and the customer experience in the years to come.

AI-centric marketing and strategic initiatives are therefore becoming increasingly important for businesses, and how effectively they can leverage AI could help determine their future growth and competitiveness.

One initiative currently attracting attention is the AI agent. AI is believed to have the potential to transform the very nature of ecommerce by gathering information and making decisions on behalf of customers, and even handling actions such as purchasing and payment.

In this article, we’ll explain the basic concepts of AI agents and how they differ from traditional generative AI and AI chatbots. We’ll also discuss the potential applications of AI agents in ecommerce, the challenges involved in implementation, and the steps ecommerce businesses will need to take moving forward.

Key takeaways

• Unlike generative AI or AI chatbots, AI agents are artificial intelligence systems that understand user intent, make autonomous decisions, and execute tasks.
  
• As Japan’s ecommerce market continues to expand, intensifying competition and rising demand for personalized experiences is driving increased interest in AI agents.
  
• Initiatives using AI agents to guide shopping experiences are gaining momentum, as seen with Amazon's Rufus, Rakuten Group's Rakuten AI, and LINE and Yahoo! JAPAN’s Agent i.
  
• AI agents are increasingly capable of assisting with the entire purchasing process—from product search and comparison to reservations and payments—and are beginning to significantly transform the shopping experience in ecommerce.
  
• The implementation of AI agents also presents challenges, such as those surrounding role of design, data integration, and security.
  

What is an AI agent?

An AI agent is an artificial intelligence system that makes autonomous decisions based on its objectives and carries out tasks accordingly. Unlike generative AI or chatbots, AI agents are distinct in that they do more than simply answer customers’ questions; they’re designed to complete specific tasks.

Imagine a customer is looking for luggage and specifies this criteria: “a front-opening carry-on suitcase in red, priced under ¥10,000, and made in Japan, with high review ratings.” An AI agent can search for products based on these criteria, compare the options that meet them, select a product, and even proceed with purchase and payment.

Thus, the defining characteristic of AI agents is their ability to seamlessly handle the entire shopping process, from interpreting requirements and selecting products to the actual purchase and payment.

The concept of agentic commerce, in which AI takes on the role of making purchasing decisions, has also been gaining attention in recent years.

How AI agents differ from generative AI

Generative AI is primarily designed to create content, such as text or images. Services such as ChatGPT and Gemini are prime examples of systems that generate content in response to user prompts.

It should be noted that many of these generative AI systems are based on large language model (LLM) technology, which is characterized by its ability to generate natural-sounding text by drawing on vast amounts of textual data.

In contrast, AI agents do more than simply generate output; they can make decisions and execute tasks based on objectives they’ve been given.

The differences between AI agents and generative AI are summarized below.

How AI agents differ from AI chatbots

An AI chatbot is a system that provides information and responds to inquiries through interactions with users. Recent advances in AI have made it possible for systems to respond flexibly based on the content of a user’s question, rather than just providing preprogrammed answers.

The primary focus of these systems is providing information in a conversational format—they do not perform any actual processing or carry out operational tasks.

In contrast, AI agents go beyond mere conversation; they understand user inquiries and take the necessary actions on their own.

For example, in response to a customer report that the product they received was damaged, the AI agent can verify the situation and proceed with the return process. Similarly, if a customer submits a request to change their order, the AI agent can update the order information directly, thereby resolving the issue.

The differences between AI chatbots and AI agents can be summarized as follows.

• AI chatbots: Provide information and guidance through conversation
  
• AI agents: Understand inquiries, determine the appropriate next steps based on the situation, and take action
  

Thus, the key distinction is that while AI chatbots focus primarily on conversation, AI agents are capable of executing actions as well.

Why AI agents are gaining attention in ecommerce

The growing interest in AI agents in the ecommerce industry stems from market expansion, intensifying competition, and heightened customer expectations.

The domestic ecommerce market in Japan continues to grow year after year. According to a survey by the Ministry of Economy, Trade and Industry, the size of the domestic B2C ecommerce market in 2024 reached approximately ¥26.1 trillion (a 5.1% increase over the previous year). The adoption rate of ecommerce stands at approximately 9.8% and is on an upward trend, indicating that the shift toward online commerce is steadily progressing.

Furthermore, the size of Japan’s B2B ecommerce market reached ¥514.4 trillion in 2024, demonstrating that the ecommerce market serves as a critical infrastructure not only for consumer transactions but also for transactions between businesses.

In addition to the market itself expanding, the number of new entrants is also increasing, making competition even more intense. As a result, it has become difficult to stand out simply by selling products, and there is now a growing demand for the personalization of ecommerce experiences.

However, there are limits to how much continuous optimization is possible when handled manually. As a result, there’s growing interest in systems that can automatically optimize the entire shopping experience.

Chat commerce—where customers purchase products through social media and messaging apps—has also been gaining in popularity, and purchasing behavior driven by conversations with customers is becoming increasingly common.

AI agents are seen as one way to address these needs and changes in the ecommerce environment.

Use cases for AI agents in ecommerce

AI agents are being used with increasing frequency in various ecommerce roles, such as customer service and marketing.

In this section we’ll introduce specific use cases based on initiatives from leading companies.

Evolution of interactive customer service through Agent i, from LINE and Yahoo! JAPAN

In recent years, LINE and Yahoo! JAPAN have begun offering an AI agent called Agent i, marking a significant step forward from traditional chat-based customer service.

Agent i is an AI agent that integrates Yahoo! JAPAN's AI Assistant and LINE's LINE AI. It can be accessed via the icon displayed on the LINE tab or beside the search box on Yahoo! JAPAN.

The goal with Agent i is not only to provide information and make suggestions through interactions with customers, but also to support processes such as searching, comparing, and decision-making, with the ultimate aim of eventually handling the execution of tasks as well.

Enhanced shopping experience through Rakuten Group's Rakuten AI

The Rakuten Group has introduced Rakuten AI, an AI agent for their Rakuten Travel service. This AI agent is actively working to support and enhance the shopping experience for customers.

When customers provide the AI with their budget and preferences, the AI agent suggests potential accommodations. The customer can then choose the plan that suits them best and proceed directly to booking.

The system can also suggest repeat bookings or similar plans based on past reservation history, allowing it to make recommendations tailored to customers’ behavior and preferences.

Conversational shopping support through Amazon's Rufus

Amazon has introduced an AI shopping assistant called Rufus that answers customer questions and recommends products.

Customers can ask questions about a product’s uses and features, and the AI provides relevant information to help them make purchasing decisions.

While it’s currently an evolution of traditional search and recommendation functions, this format, which supports purchasing through dialogue, can be characterized as an AI agent-based approach.

Challenges in implementing AI agents and how ecommerce businesses can overcome them

While AI agents are excellent at enhancing the customer experience on ecommerce sites and improving operational efficiency, there are several challenges to consider when implementing them. Let’s take a closer look at these issues and how they can be addressed.

Designing the role for AI agents

While AI agents can be used for a wide range of tasks, one challenge is determining exactly how much responsibility to entrust to them. The design requirements and risks involved vary significantly depending on whether the system is intended solely to suggest products or whether it’s also responsible for handling actions such as purchases and payments.

Therefore, when implementing AI, it’s important to clearly distinguish between tasks to be handled by AI and those to be performed by humans, and to explicitly define the roles and responsibilities.

Strengthening data and system integration

For AI agents to function properly, it’s necessary for them to be properly integrated with product information, inventory, and customer data. If integration with existing systems or data organization is inadequate, it could compromise the accuracy of recommendations and the consistency of processing.

To address this, it’s necessary to centralize data management, strengthen system integration, and establish an environment in which AI can access the necessary information in real time.

Establishing a secure payment environment

When AI agents are involved in the latter stages of the purchasing process, ensuring payment security and mitigating the risk of fraud become key challenges. In particular, situations involving the handling of customers’ personal and payment information require a high level of security and reliability.

It’s therefore important to implement measures to prevent fraud and ensure security, while also designing a smooth payment process that allows users to complete their transactions with confidence.

For ecommerce businesses moving forward, a key priority will be creating an environment where AI identifies and recommends products, and customers can then make a decision and complete the payment process safely and smoothly.

How Stripe Payments can help

Stripe Payments provides a unified, global payments solution that helps any business—from scaling startups to global enterprises—accept payments online, in person, and around the world.

Stripe Payments can help you:

• Optimize your checkout experience: Create a frictionless customer experience and save thousands of engineering hours with prebuilt payment UIs, access to 125+ payment methods, and Link, a wallet built by Stripe.
  
• Expand to new markets faster: Reach customers worldwide and reduce the complexity and cost of multicurrency management with cross-border payment options, available in 195 countries across 135+ currencies.
  
• Unify payments in person and online: Build a unified commerce experience across online and in-person channels to personalize interactions, reward loyalty, and grow revenue.
  
• Improve payments performance: Increase revenue with a range of customizable, easy-to-configure payment tools, including no-code fraud protection and advanced capabilities to improve authorization rates.
  
• Move faster with a flexible, reliable platform for growth: Build on a platform designed to scale with you, with 99.999% historical uptime and industry-leading reliability.
  

Learn more about how Stripe Payments can power your online and in-person payments, or get started today.