First, the background: The future of artificial intelligence arrived with a bang this spring in the form of Microsoft’s rollout of its new and improved search engine, Bing, along with a chat function that lets users converse, debate, and even flirt with the Internet. Early reviewers say it may be the most significant technological development since the iPhone and that, like the iPhone, it will drive a surge in innovation, productivity, and disruption.
The feature that’s creating the most buzz is called ChatGPT, a natural language processing algorithm that responds to prompts (questions) like an online customer service chatbot, except this one is on steroids.
Instead of a list of ranked answers to a typical Google search, ChatGPT replies with a grammatically perfect text answer with links to sources.
ChatGPT will compose emails for your specific needs and write college essays and business reports. It’s being used on websites to write articles based on information scraped from the Internet. Watch for the absence of a human byline on a story like this one, and you’re likely reading a story generated by an algorithm.
It’s become such a cultural phenomenon in such a short time that it was the subject of a brilliant spoof on Comedy Central’s South Park in a recent episode that does a surprisingly good job of explaining how ChatGPT works and what’s flawed about it. A recent Intelligent.com survey of 1,000 college students found that about a third have used ChatGPT to burnish their essays and reports, even though most think it’s cheating.
The retail industry is a prime candidate for practical applications, as I found when I tested ChatGPT with this question: What impact will ChatGPT have on the retail industry?
The near-instantaneous reply:
As an AI language model, ChatGPT can be integrated into communication channels for retailers to interact with their customers in a more efficient way. Its impact on the retail industry can be summarized as follows:
1. Better customer experience: ChatGPT can be used to improve customer service and provide personalized shopping assistance. By understanding customer queries and responding with appropriate answers, it can reduce waiting times and enhance the overall shopping experience.
2. Improved customer engagement: Retailers can use ChatGPT to engage customers on social media platforms and websites. It can be used to answer product-related questions, suggest product recommendations based on browsing history, and even promote new products.
3. Reduced costs: ChatGPT can automate routine tasks and reduce the need for human customer service representatives. This can help retailers save costs on hiring and training staff.
4. Increased sales: With personalized recommendations and 24/7 availability, ChatGPT can increase sales by providing customers with a seamless and convenient shopping experience.
Overall, integrating ChatGPT into the retail industry can lead to improved customer satisfaction, increased sales, and reduced costs.
It all sounds terrific and entirely believable — someday. For now, the excitement seems to be focused on ironing out the many kinks in software development and determining how to make money with it. Challenges abound.
For starters, ChatGPT was trained on content that existed on the internet up to the year 2021. We asked a ChatGPT-related search engine called Phind, “Why did Bed Bath & Beyond fail?” The answers parroted the coverage on CNN and selected investing sites, with links to nearly three years old articles.
Critics say ChatGPT is factually unreliable and shallow. Many answers solicited by people using the system contained errors, and responses tended to lack depth or insight.
It will take retailers longer than most industries to reorganize their data management to take advantage of the new capability. The industry is notoriously stingy and slow in adopting new technologies.
Finally, ChatGPT and AI are expected to produce turmoil in retail, brand advertising, and marketing. Currently, an AI-generated text reply to a search question does not use the traditional ranking system that Google searches rely on. A significant unknown is how these human-like robo-conversations will affect ad revenue.
For now, said Microsoft CEO Satya Nadella, “We’re starting with how to reshape the largest software category on planet earth: search.”
There is clearly work to be done to make it not only efficient but effective.