“I don’t understand this at all.”
“I hate it.”
“Why can’t we go back to what we had before?”
If these are familiar reactions when someone mentions Artificial Intelligence (AI), you’re not alone. Many people think it’s evil and an evolution of technology personified by a hard, physical construction like a robot. Countless science fiction movies have ingrained into us the idea that robots will be spewing out intelligent facts and information, perfectly mimicking human speech, behaviour and intelligence. Humans will be redundant. This isn’t true.
Sceptical? Or curious?
Well, to recognise AI and its potential to augment your life’s experiences, the story begins with data and goes back many years.
Remember those times when you bought something and the merchant asked you for details like your name and address? Months later, you’d receive a catalogue for this merchant’s newest products. Well, that probably meant that someone collected this data, processed (entered it into an Excel sheet) and analysed it. The outcome was that you’d be a suitable person to receive the merchant’s next advertisement. It didn’t matter that what you first bought was a tea pot and the merchant was now advertising an oven. This was loosely referred to as structured data within the realm of what was called Business Intelligence.
Fast forward a few years and there was an explosion of data collected. In fact, it’s still happening. Currently, we have collected 8 zettabytes (according to Wikipedia, a zettabyte is a measure of storage capacity and is 2 to the 70th power of bytes). By the year 2020, we are predicted to have close to 40 zettabytes of data collected.
All this is called Big Data and it includes both structured and unstructured data. Unstructured data includes information that is collected from social media, emails, websites, online shopping transactions. Keep in mind that you’re a living, breathing generator of data. When you search the Web, send an email, comment on social media or shop online, systems somewhere out there are collecting the ‘data’ you have created.
This colossal amount of data is no longer collected by humans who then enter such information into an Excel sheet. It’s collected automatically by robots.
Next comes trying to figure out what to do with all this data. If none of that data is properly analysed, you’ll still receive an advertisement for an oven.
This is where Machine Learning and Deep Learning come into play. Humans create programming language where these machines can look at all this data and identify nuances and patterns in human behaviour. Once they’ve analysed this, they are able to tell a merchant what you are more likely to be interested in. You may now receive an advertisement for similar tea pots. What makes it fascinating is that with automation, this entire process can happen in a short time. It won’t take months for merchants to figure out what would suit you.
Remember the time you bought a bottle of chilli sauce online? Next time you’re online, observe how advertisements for other brands of chilli sauce pop up on your screen. You may even receive information that another brand is better because its sugar content is lower and can help you manage your diabetes.
Because of your song playlist, your electronic voice assistant, Alexa, will play your favourite song to wake you up in the morning. Since you drive to work, traffic and weather conditions will be monitored by your GPS and it’ll automatically suggest your shortest and fastest route. As such, AI already manages how you behave because you’ll be presented with timely, relevant and useful information that is tailored to suit your needs and wants. In other words, AI augments your human experience.
FOLLOWING THE RULES
A significant component of AI is what’s called a ‘rules engine’. This is built into the machine which controls how it works without human intervention. Take the classic example of a racing car in the Formula One industry. It typically has over 300 sensors to collect over 50,000 unique channels of unstructured data like outside temperature, air pressure and humidity. Using graphs, relationships between all sorts of data points can be visualised.
When weather conditions are not optimum, a warning light goes off. The sensors collect the data that ‘informs’ the car’s rules engine. This kickstarts the process of ‘correcting’ conditions ie. reducing speed to reduce the machine’s temperature. These will allow the car to better perform during a race or indicate the need for timely maintenance of parts.
Moreover, such visualisations are not restricted to the present. AI allows you to take unstructured data like weather forecast for ten years from now, correlate it with data about engine performance and come up with simulations to ensure that the engine of tomorrow is far superior to what you have today. That’s the power of AI.
It is worth repeating that AI isn’t about better programming or programming better machines. Instead it is about trying to understand our behavioural patterns so that our lives are augmented.
For a merchant, the purpose remains the same which is to capture his customers’ attention, invoke their emotions and ensure they buy his products.
As a customer though, you’ll avoid wasting time and money. You’ll know that those tea spoons you bought match your set of tea cups and tea pot. You won’t have the burden of keeping an oven that you bought on an impulse.
Basically, AI is no longer merely an evolution of technology, but an ecosystem of different technologies that work towards a smart outcome. We must accept that AI is here to stay.