ChatGPT is here to stay. How small businesses should use it to their advantage

If you ever worked with Office 97, you’ll remember Microsoft’s iconic assistant Clippy. The animated AI-based paperclip would pop up in your document and offer to help you format whatever you were working on. Clippy would analyse and structure words in your document and help you format it according to what you were working on. For example, if anything vaguely resembled a resume, you would be prompted with “It looks like you’re writing a resume! Perhaps a resume template will help?”. 

In 2023, Clippy’s legacy lives on in our everyday lives in the form of much more advanced personal digital assistants powered by emerging techniques in artificial intelligence (AI). The latest example is OpenAI’s ChatGPT, which has led to another resurgence of mass public excitement and debate about AI.  Some are even calling it a watershed moment for society, as significant as the advent of the internet.

For small businesses who are already comfortable in the digital realm and happily making use of time-saving features of many products that are powered by AI, the public release of ChatGPT and associated rollouts has sparked excitement over the possibilities it unlocks. For others yet to dip their toes into the world of human-machine augmentation and AI, you may have FOMO or even be fearful.

While ChatGPT, Google’s Bard, Microsoft’s AI-enabled Bing and the many other fascinating applications built on today’s generative AI technologies are a way station along the road as AI continues to evolve at pace, more change is definitely coming. AI techniques hitting prime today will have broad implications for the type and manner of work conducted by small businesses in the near future. 

It’s vital to understand its uses and limitations now, to set your business up for success.  

What actually is ChatGPT

ChatGPT belongs to a class of AI techniques known as generative modelling. It’s trained to mimic the complex statistical relationships that exist between words in a language, hence the technical name of Large Language Models or LLMs. Using these learned statistical relationships, LLMs produce responses to a very wide range of questions that look like they could have been written by a human.

LLMs can communicate in plain English, generate and edit text, and write code. In doing so, it opens up the potential for small businesses to produce more work, faster than ever before. For example, ChatGPT’s ability to do all different kinds of writing like generating boilerplate text for proposals, social media or blogs could prove useful for small businesses — particularly those less fluent in the dominant language in their market or less comfortable with creative or technical writing.

Why all the buzz now

For many professionals working in AI, generative AI is not new news. What makes ChatGPT exciting is not the underlying models themselves but the simple ‘human accessible’ interface and ‘free for now’ wide availability. You don’t have to be an AI expert or a programmer to play around with the amazing capabilities offered by ChatGPT and that’s been a game changer in capturing the public imagination. 

While LLMs have been around for a while, the end-to-end ChatGPT user experience (which, critically, includes an intuitive, easy-to-use user interface) has changed the game. ChatGPT is not without its limitations, as many enthusiastic tinkerers have discovered. As with most new products, many aspects need further development. Already, the latest version of ChatGPT released last week is proving to be “more reliable, more creative” and able to handle more nuanced instructions than the previous version.  

So, what does it mean for my business

Over the coming weeks and months, we can expect to see a wave of new businesses and applications built on top of generative AI models like ChatGPT. How many of these applications will have a commercially robust business model once the hype and investment subsidies? Time will tell. 

To distinguish between the useful, the shonky and the revolutionary, small businesses should approach ChatGPT in the following ways:

Experiment, but with caution 

The best place to start is experimentation. There is value to be had in playing around with the tool to test its knowledge, its limitations and boundaries. First-hand experience with the tool can help you understand how it could be used in your business and the wider industry. But a word of warning: as with all technology that is free and open, if you’re not paying for the product, you are the product. And the usage model (free to use right now) could change at any time and without warning.

Think of it as a business tool

One way to think about ChatGPT is as a tool alongside the many other tools you use every day in your business. For some small businesses, ChatGPT and other language models may seamlessly integrate into their existing business platforms, and work behind the scenes to speed up existing processes. It will likely integrate into digital experiences they’re already familiar with, like generating reports. 

But remember, just like you need many tools to build a cabinet, ChatGPT should be used in conjunction with other business tools — it can’t be relied upon in isolation.

Analyse the likely impact

ChatGPT has the potential to create new business models, jobs and workflows. Small businesses should consider the tools and processes at the heart of their business and how AI technology could assist. For example, automating customer service and responding to customer requests. This means understanding not only how the technology could be used in chatbots, SMS, and email for personalised messages, but how AI could free up your teams to focus on improving the customer experience.

We’re just at the beginning of this AI epoch and there’s no telling for sure where it will lead us. By being open to adapting and experimenting, arming yourself with knowledge and working with your teams and advisors to decode its impact and opportunities, you’ll be better placed to incorporate AI into your business in a way that genuinely adds value to your bottom line. 

Dr Kendra Vant is the Executive General Manager of Data at Xero. With doctoral research in experimental quantum physics at MIT and postdoctoral work in applied quantum computing at Los Alamos National Laboratory, Kendra has worked in New Zealand, Australia, the US and Malaysia, and has more than 15 years experience leading data and technology teams with a particular focus in FinTech.