It wasn't that long ago that artificial intelligence was seen as a luxury reserved for large corporations; a promising technology that was just out of reach for small and medium sized businesses (SMBs).
Those days, however, are gone. Today, there are a number of inexpensive, and sometimes free, tools available to small business owners. From chatbots to software that manages your cashflow or protects online stores from hackers, AI can make a difference for business owners looking to automate customer service, streamline operations and enhance customer experience to drive growth.
Jeff Chan, Senior Manager of Enterprise Innovation at TD, leads a team that explores the possible role emerging tech (like generative AI) can play in meeting future customer needs. Generative AI – a subfield of artificial intelligence that powers the OpenAI chatbot, ChatGPT – describes algorithms that can be used to create new content, including audio, code, images, text, simulations and videos.
Chan said that for many small business owners, the benefits of AI are already available and will continue to improve over time.
"Degrees of automation and automated decision making are already being used by small businesses today to help them manage inventory, follow up with clients and do things like improve their research and development practices," he said.
"Thanks to the explosive growth of generative AI, I believe that how we currently operate, think and analyze data will change in ways we thought were still years off."
Of course, Chan is quick to point out that Generative AI is in its early stages and there are risks associated with the use of the emerging technology of which small business owners need to be aware. Small business owners must assess these risks, as well as the terms governing the use of third-party AI systems, in association with their own use cases, before implementing any new AI system, including Generative AI.
Here are three areas where small business owners may benefit from integrating AI into their businesses, provided the risks are understood:
Delivering human-like customer service at scale
Chan said that tools like automated responses have established themselves as staples of the customer service experience. One such common technology, Interactive Voice Response (IVR), allows telephone users to interact with a computer-operated telephone system through the use of voice or a telephone keypad.
The purpose of automation is to free up time by prioritizing, managing, responding to and resolving customer inquiries to help improve the customer experience. Today's chatbots are limited in what they can resolve.
According to Chan, this is because it's extremely difficult to train those chatbots to understand human, or natural, language. But chatbots powered by generative AI, like OpenAI's ChatGPT, are able to understand conversational language.
"Having that ability could unlock new opportunities to help resolve customer service issues in a more human-like way and not through pressing buttons," Chan said.
"The more likely it can be resolved by an artificial intelligence, the more time you give human service team to focus on complex issues that would benefit from more human involvement."
As AI continues to improve, interactions by customers with anything digital will begin to sound and feel more and more human-like, which is the goal when you're looking to create a more personalized experience, Chan said.
New insights and customer experiences from existing data
Because AI can rapidly process massive amounts of information, it has the ability to comb through an organization's data to identify patterns and make connections that might otherwise go unnoticed. Chan said that many small business owners may not realize that they're already using AI to optimize their businesses.
"AI-driven solutions are no longer reserved for companies with deep pockets," he said.
"Now, there are very cost-effective solutions that can give small businesses the power to do more with their data than ever before. Next generation AI should be affordable, giving entrepreneurs more ways to find those nuggets of insight that could help them uncover net-new revenue opportunities."
At the same time, small business owners looking to ensure highly-personalized experiences for their customers can use AI to help improve their messaging based on their customers' preferences and behaviours.
However, there are some critical considerations small business owners need to think about when implementing AI technologies, including having an understanding of the limitations and assumptions of any proposed technology, Chan said.
Small business owners should also assess the terms governing the use of any third-party AI technology, including for example, data impacts, privacy risks, and intellectual property implications.
For example, Microsoft on its website advises users of its Azure Open AI models to "carefully consider well-scoped chatbot scenarios."
"In other words, if you're going to use AI for customer service, focus on scenarios that are straightforward like issuing a refund or checking the status of a package," Chan said.
Limiting the use of chatbots to a narrow domain may help to reduce the risk of generating unintended or undesirable responses, but each technology and use case must be assessed individually.
Fraud detection and cybersecurity
Cyber threats and fraud remain a big threat for businesses of all sizes. According to credit reporting organization TransUnion, digital fraud attempts in Canada have spiked 189% from pre-pandemic levels. At the same time, the Canadian Anti-Fraud Centre received cybercrime reports totaling $530-million in victim losses in 2022.
Businesses are increasingly reliant on AI to help protect them and their customers from fraud. Chan points out that AI-powered security can flag suspicious activities and unusual patterns in real time, which can then alert employees and small business owners to take action.
As a final thought, Chan emphasized that while AI on its own can be a helpful technology, it still requires human intervention to be truly effective and to help reduce some of the risks of associated with a still-nascent technology.
"In my opinion, nothing can be 100% automated," Chan said.
"At some point, a human needs to step in. Digital alone isn't enough. Human alone isn't enough. I believe a digital-human hybrid experience will enable the best type of customer experience."