AI or Us… Or Both?
AI is changing the way we work, the way we learn, the way we live. There’s no debating that. It’s just a fact. What can (and rightfully should) be debated is the question of will it replace us.
It’s a huge question and so many people are throwing so many different answers out there about it. So, instead of just throwing my answer on top of the pile, I’m going to go the much more fun route of not giving a straight answer (because otherwise it’d just be too easy!). I’ll give my thought process and a few places to find out more information, and let you decide for yourself where you think AI is going.
What is AI?
The field of Artificial Intelligence has come a long way over the past few years. There’s no universally agreed definition for what an AI is but there are two common (but quite different) definitions and their differences.
The first ‘AI’ is, what we call a “General Artificial Intelligence”. This is a machine or computer that can think like a human does and make decisions like we do. It can hold conversations and pass the famous “Turing Test”. This is the form AI normally takes in science fiction, like HAL 9000 in 2001: A Space Odyssey, the T800 in Terminator or Ultron and Vision in the Marvel Cinematic Universe. They are ‘human-like’.
The second ‘AI’ we have is the more realistic view of modern-day technology. This is the type of AI that we use in our lives already. Machines are exceptional at finding incredibly complex or hidden patterns in data. They can discover things we never could have but they don’t understand “the why” behind these patterns. They make a decision based on whatever statistical model it was built with. But they don't have initiative or creativity. You could describe them as ‘smart but not clever’.
So what are humans better at?
A lot of things. We can be creative and empathetic, unlike machines who can at best try to imitate that. For example, say we wanted to make an ad campaign for a new movie. An AI could look at all the data from past ad campaigns and see patterns to find what works. It could find that Youtube ads are the best way to advertise a comedy but TV ads are better for action movies.
But that’s not anything new. That will mean your ad campaign will be just like the ones before it. That’s not to say that it won’t be effective, but following that pattern will never lead to a new strategy. Whereas a human could think of something novel; a risk. Something that could lead to your movie being the talk of the world for a while… or something that could lead to box office failure if it doesn’t work.
Take the marketing for Deadpool in 2016 as the perfect example of a brand new campaign that got the whole world talking about it. An AI could never have done that. At least not any current AI, maybe in 20/30 years.
Why Use AI At All Then?
For a few reasons, but the key one: not every decision we make needs to be a completely unique, brand new idea. Trying to decide what stock you need to order for your shop is nearly always dominated by future projections that are built from past sales. Most shops sell the same products week to week, month to month, year to year. So an AI can find the patterns in sales and be ready to follow these trends over time.
If you have a database up and running, you don’t need to be constantly finding new ways to design and rebuild it over and over again. You build it well the first time (or at least should!) and then keep up the maintenance as you go along. Take the Autonomous Database, for example, it uses AI to structure, maintain and secure itself on a level humans could only hope to match. Some things machines are just better at.
AI also doesn’t (or at least shouldn’t) hold our same conscious or unconscious biases. If you wanted to find the best candidate for a job, AI can help you sift through applicants in a fairer way than a recruiter might be able to. But only if it’s developed and trained correctly. If used incorrectly, AI could actually create a less fair system. It could learn our biases from data we give it. And in that case, who is responsible?
So what do you think? Would it be better to move to AI taking over from humans? Or would we be better off not taking any risk and just ignoring AI altogether, letting it become a forgotten thought from the past?
Maybe we’re better off together? Maybe that’s where we were always going.
Final Note
Recently I had the pleasure of working with Yusuf Jazakallah of RecruitmentSmart and Riham Satti of MeVitae on an episode of the CurioCity Podcast about Artificial Intelligence (Season 2, Episode 2) and how it can work alongside people in the workplace. It was really interesting to get their perspectives on where the field is going and I would highly recommend giving it a listen if you’re at all interested in the area.
You can find it on Spotify, YouTube or iTunes now. Make sure to let us know what you think about our future with AI!
And if the thought of the Autonomous Database interested you at all, you should definitely check out one of the free webinars coming up about the Autonomous Database, no previous knowledge needed! There’s one on the 27th February and one on the 26th March so make sure to check one out!
