Digiterre’s Rajesh Jethwa is named in the top 30 of the 2023 CIO 100
Aren’t you tired of reading about self-driving cars and deliveries that turn up on your doorstep before you’ve even realised you need them? It’s all so futuristic. When is all this Artificial Intelligence (AI) talk going to turn into something of practical use to your business?
Well, it already is. AI is at work in numerous everyday technologies, like email filtering and internet searches. If you’re hoping for something a bit more spectacular than that, it could be that your definition of AI has been exaggerated. ‘Artificial Intelligence’ is a term first used in 1955 to describe machines that are capable of intelligent behaviour. In its basic form it requires human programmers to give the machine a set of parameters, which it can then use to categorise information, make decisions, provide answers and so on.
But by the time AI entered the vernacular, scientists were already exploring the possibility of creating machines that could learn for themselves. The science hinged on the principle that the human brain learns from experience, by gathering and categorising observations, which then provide the basis for future decisions. If they could build an artificial ‘neural network’ that could mimic the workings of the human brain, they could then plug it into an information source and leave it to learn for itself.
For decades the principle delivered little more than a series of computers that were good at chess, but then along came the internet – the ultimate information source – and ‘Machine Learning’ (ML) hit the mainstream as an application of AI. A good way to think of the difference between AI and ML is: AI is the science for making the machines and ML is the algorithms that make the machines smarter. Whilst ML is still small scale, it’s growing really fast.
Here are five applications of both that could help your business run more efficiently and deliver a higher level of service.
The more you understand about your customers and competitors, the more effectively you can deliver your products or services. AI can analyse vast amounts of data from a range of sources, such as internet use, EPOS data and even social media, to build in-depth insights into customer needs and behaviours, enabling you to predict and meet demand.
Amazon is the headline grabber for this, and its plans to deliver us goods it knows we need before we do are bound to come to fruition before long, but supermarkets have been using AI in this way for years too. They target each customer’s learned preferences with bespoke coupons, discounts and ads, and even stock their shelves according to the customer trends they’ve picked up through AI. Think about how this could help your business. It means you no longer have to treat your customers as ‘one’ but can target each one individually with an accuracy and efficiency that would have been impossible without AI.
Customer service is being revolutionised by AI and ML, which is entering the realms of artificial emotional intelligence. More and more companies are using chatbots, which give the impression of a human at the end of the line but are actually autonomous machines, programmed to handle online enquiries, gather information from the enquiry, recommend solutions and report back to the business, including notifying a member of staff if human action is required.
Chatbots are becoming increasingly good at responding as a human operator would and now ‘cognitive AI’ programmes are giving the machines empathy. Human customer service staff can’t always be relied on to judge the appropriate tone to use with customers but cognitive AI can help them. Cognitive AI programmes listen to customer service calls and analyse the customer’s mood from their voice and other traits. They can then provide live guidance to the human operator to help them speak to the customer in a way that will get the best response.
It doesn’t matter what business you’re in, these days you need to include a publishing function amongst your services. Digital marketing depends on content, which means hiring writers to produce it for you. Lots of it. But what if you could programme a machine to produce content for you? At the risk of putting myself out of work, I can tell you it’s possible.
News agencies are using AI to write small, data driven items, such as sports stories and financial summaries. Happily for me, AI hasn’t mastered the art of producing beautifully crafted articles like this one, but give it access to a set of data and it can interpret the figures to pull out a relevant headline or set of bullet points that summarise the insights within. Along the lines of the market intelligence technology mentioned above, automated marketing platforms are also using AI to create targeted content and ads, learning how and where your market plays and generating bespoke content to engage them, without any human intervention. Automated marketing is now being used by more than half of all businesses, in some way.
Giving a machine the ability to understand speech is pretty smart and businesses have been using voice recognition software for years to speed up processes like letter writing, but now the machines don’t just understand what you’re saying, they can recognise your voice, even down to the tone you’re using and what that might mean. Virtual Assistant (VA) software, of the sort that drives Siri and Alexa, could provide you with the perfect PA. VAs can take a letter, manage your diary, arrange appointments, send invitations and sort your emails, all in the blink of an eye. And if they can’t actually bring you a cup of coffee, they can at least tell you the best place to get one.
Picture a security guard sitting in front of a bank of monitors, each one relaying pictures from a different part of your premises. You’d like to think they’re keeping a constant watch, always ready to react if something suspicious appears on one of the screens, but you know they’re bound to glance at their newspaper from time to time, or pour a cup of coffee from a flask. Some security guards have been known to fall asleep.
Thanks to ML, you can train a machine to monitor your surveillance every single second of every day, without ever tiring or being distracted. In the same way that AI can be programmed to look for anomalies in banking activities to flag up potentially fraudulent transactions, machines can be taught to watch for signs of unusual behaviour on CCTV. By working in conjunction with humans, the technology can make traditional surveillance much more robust. In a similar way, AI and ML can be used to protect against cybercrime too, scanning your networks for unusual activity, learning what may or may not constitute a threat and taking timely action to shut down systems or seek and destroy malware without needing human intervention.
The way businesses work with AI is becoming increasingly important. There’s still a lot of work to do before we’re all being chauffeured around in driverless cars but we can be reaping the benefits of AI and ML right now. Choosing not to integrate AI into your workforce is akin to forcing your staff to work with knitting needles when all your competitors have the spinning Jenny. Not only does it improve the speed and accuracy of mundane, repetitive processes, it can work alongside your human workforce to effect unprecedented levels of performance too.
It’s time to embrace the AI future.
By: Rajesh Jethwa
Get the latest news and stay up to date