Michelle Xiang 
“AI is likely to be either the best or worst thing to happen to humanity.” — Stephen Hawking
Since the advent of AI, opinions on its potential in various fields have been divided. This division is especially apparent in the realm of AI-powered e-commerce customer service. Many hold misconceptions about the capabilities and limitations of AI in this context, such as concerns about complexity and AI’s ability to provide relevant answers.
Today, we want to dispel some common misconceptions surrounding the use of AI in e-commerce customer service.
Misconception 1: AI Will Replace Human Customer Service Agents
A report by the World Economic Forum suggests that by 2025, 85 million jobs will be replaced by artificial intelligence machines. However, will customer service jobs be replaced? I believe the current answer is no.
While AI-powered chatbots can provide instant support and effectively handle common issues, AI has not yet been able to replicate the depth of human interaction and understanding. It lacks the ability to provide the empathy, adapt to unique situations, and offer emotional support that humans can provide.
Rather than replacing jobs, AI can complement human efforts by streamlining operations and improving efficiency. This allows human agents to focus on more complex and nuanced interaction. Unlike other AI chatbots, Cheerlink’s Ama AI chatbot only engages in conversations during the time frames you allow. If a customer service agent gets involved in the conversation, Ama’s switch will be turned off, allowing the agent to take over.
The future of customer service lies in the harmonious coexistence of AI and human agents, delivering a superior and more efficient customer experience.
Misconception 2: AI-Powered Customer Service Lacks Personalization
Some skeptics argue that AI is too rigid and formulaic to cater to individual preferences and needs. However, this perception does not align with the capabilities of modern AI systems.
For example, the AI chatbot in the customer support tool Cheerlink is powered by large language models (LLMs), and OpenAI’s latest GPT-4. It can comprehend the context of user queries, allowing for more natural, context-aware conversations with users.
Moreover, Cheerlink’s Ama AI chatbot will soon be able to recommend products based on a user’s browsing and purchase history. This level of personalization can not only enhance user satisfaction but also drive sales growth.
Misconception 3: AI-Generated Responses are Unreliable
Many people have reservations about the answers generated by AI chatbots. This is because some AI chatbots, while seemingly providing reasonable and coherent information when answering questions, this information is entirely inconsistent with reality, and sometimes even fabricated — a phenomenon referred to as “AI illusions.”
For developers of customer service AI chatbots, this is a critical issue to avoid. For instance, AI chatbots should avoid recommending products that don’t exist in the store and refrain from making erroneous statements about discount policies to prevent transactional confusion.
Cheerlink accomplishes this. Firstly, Cheerlink’s Ama AI support filters out irrelevant (“toxic”) questions before responding and doesn’t answer questions unrelated to your store. Secondly, it generates responses based on your selected content, such as product information and FAQs.
Misconception 4: Customers Aren’t Ready for AI Chatbots
The reasons people harbor negative sentiments towards traditional chatbots are:
Traditional chatbots hinder users from connecting with real human agents who can effectively resolve issues. But AI chatbots like Cheerlink’s Ama AI can offer a “handoff to human” feature (coming soon). When your customers’ questions go unanswered, they can choose to switch to a human agent, making it very convenient.
Limited natural language understanding abilities of chatbots result in poor user experiences — users need to use keywords to trigger the correct responses from the chatbot. However, the emergence of large language models (LLMs) and OpenAI’s latest GPT-4 seems to be changing the landscape. AI chatbots supported by these technologies can engage with users in a more natural manner, making users feel like they are conversing with real humans. Confidence in their potential is high, as stated in a report by Intercom: “73% of support leaders believe that customers will expect AI-assisted customer service in the next five years.”
Misconception 5: Setting up AI in E-commerce Customer Support is Overly Complex
Modern AI solutions are designed to be user-friendly and accessible, and Cheerlink’s AI customer service functionality is a prime example of this. Cheerlink empowers Shopify merchants to easily select the information AI customer service should reference when responding, all without requiring extensive technical expertise. It’s important to keep in mind that AI simplifies rather than complicates the customer service experience.
In conclusion, while misconceptions persist about AI in e-commerce customer service, it has become a powerful tool for improving efficiency and enhancing the customer experience. Properly understood and configured, AI customer service systems can help businesses better meet customer needs and gain a competitive edge. By dispelling these misconceptions, we can embrace the future of AI in e-commerce customer service with confidence.
