Let’s talk chatbots

Once upon a time, messaging apps such as Facebook Messenger, WhatsApp, Slack and Viber did exactly what they said on the tin: provide a platform for the exchange of messages between users.  However, recent advancements in artificial intelligence and machine learning have fuelled the developments of chatbots, which have been adopted by many such messaging platforms. These apps have now overtaken social media in popularity, with WhatsApp boasting over one billion users per month; a seventh of the world’s population. This year, as machine learning and natural language processing reaches new capabilities, bots will likely play a key role in the communications strategies of a growing number of businesses.

Tech giants say chatbots are ‘the new apps’, but what exactly are they? Think of Amazon’s Alexa or Apple’s Siri and you’re on the right lines. A chatbot is an automated program, powered by rules and often incorporating AI or machine learning, which can mimic conversations and interact with users. They now come in all different forms depending on their purpose, whether it’s ordering a taxi, reporting the weather, ordering groceries, scheduling a meeting or aiding a user with a purchase. The Chinese chatbot Xiaoice, developed by Microsoft in 2014 and now used by over 40 million people, even acts a companion for many of its users. The chatbot remembers previous conversations and asks how users are feeling, makes jokes and even uses emojis. According to figures from Microsoft, the average user who adds Xiaolce talks to her more than 60 times per month.

Chatbots aren’t actually a new technology. Pioneers in the field have been experimenting with machine learning and language processing since the 1950s. But it’s with the recent proliferation of messaging apps that the extensive potential of bots has become clear.

Chatbots can offer instant access to a company’s products or services, making the technology ideal for businesses looking to streamline communication and improve customer engagement through service automation. Over the past year we’ve seen a number of brands jump on the chatbot bandwagon. Domino’s Pizza, Burberry, Uber, Sephora and even the Guardian have all launched chatbot services, and this year it’s likely we’ll see even more companies adopt the technology.

This is not to say that chatbot technology hasn’t experienced teething problems. You only need to look at Tay: launched by Microsoft last year, the bot made headlines after Twitter trained it to become a misogynistic racist in less than 24 hours. A recent survey found that 70% of Facebook’s chatbots don’t understand their users, with only 30% able to fulfill requests without a human stepping in. As a result, Facebook is set to scale back its use of AI-based chatbots, and has issued new guidelines for chatbot developers which indicate a shift in the company’s strategy.

So, what does the future hold for chatbots? The technology is certainly one of the most discussed divisions of AI right now. As language processing and machine learning advances, it is hoped that bots will become smarter and better understood, and will ultimately be integrated into a wider range of technologies. Estimates predict that chatbots will become commonplace in less than five years. A recent survey by Oracle, for example, found that 80% of businesses already use or plan to use chatbots by 2020. Gartner also forecasts that intelligent automation will manage 85% of businesses’ customer relationships within the same time frame.

The overarching question is whether the technology behind chatbots can live up to the expectations of users, and whether consumers will entrust chatbots to deliver the same level of customer service and interaction as their human counterparts. The bot revolution has only just begun and it’ll be interesting to see whether the technology will be merely a passing fad, or will be the talk of the town for years to come.