Over the last few years, there’s been a whole lot of hoopla about Artificial Intelligence (AI) taking over people’s jobs, potentially leaving hundreds out of work. It sounded like mere speculation into the future, but in reality it’s not even ‘just around the corner’ any more – it’s here. According to a Mckinsey report, all companies should be preparing for the digital wave of disruption AI is bringing:
The impact on jobs may sound scary, but it doesn’t need to be. AI won’t only result in a loss of jobs, but the creation of new work. But how did AI emerge? Is it really all it’s made out to be? And will AI cause dramatic changes in 2018?
The Emergence of the Machines
Employing AI to automate repetitive and mundane tasks makes sense in a few ways – it’s cheaper labour for businesses and gives people the chance to do more rewarding and creative work. AI is being incorporated into the machines developed by mechanical engineering experts to help automate certain processes in the industrial and business sectors. The benefits are highlighted on a largest scale by Amazon. As well has launching their smart assistant Alexa, Amazon employ more than 100,000 robots in their warehouses to automate manual lifting work once handled by humans.
The New York Times reported that bots not only make warehouse work less tedious for people, but also create efficiency gains that “let a customer order dental floss after breakfast and receive it before dinner”. Essentially, machines enable humans to do better and more enjoyable tasks.
This transition to a robotic workforce hasn’t appeared out of nowhere either. There has been a trend in growth:
- In 2015, 15% of enterprise companies were already using AI to automate manual, repetitive tasks. (Narrative Science)
- By 2016, 38% of enterprises adopted AI (26% for automating repetitive tasks). (Narrative Science)
- In 2017, 80% of enterprises adopted some form of AI in production (Forbes)
Whilst AI is still at an early stage of development overall, it could contribute up to $15.7 trillion to the global economy by 2030 (according to PwC’s report). And this first wave of adoption no doubt brings us right up to the doorstep of some exciting opportunities through which we can start tapping into AI’s real potential.
Without trying to be too dramatic, as tech writer and TED speaker Tim Urban illustrated, we’re reaching the verge of a new paradigm that will sweep the world:
What Tim illustrates (diagram 1) was part of a larger concept which he coined the ‘S-curve’ (diagram 2). That was published back in 2015 at the time we first started to see a rise in adoption of AI in enterprise businesses. Now it’s 2018, so let’s re-visit his theory to see what the S-curve is and it means for today:
The S-curve: Why 2018?
In his article back in 2015, ‘The AI Revolution’, Urban observed that an ‘S-curve’ is formed every time waves of progress driven by new paradigms sweep the world. Zooming out from the situation presented above, this chart shows the bigger picture:
As Urban explained, the curve goes through three phases:
1. Slow growth (the early phase of exponential growth)
2. Rapid growth (the late, explosive phase of exponential growth)
3. A leveling off as the particular paradigm matures
For 2018, AI may just be at the tail end of the slow growth phase, as it’s more ready than ever to accelerate after the high adoption rates we saw over the past year. Just like growth of the Internet between 1995 and 2007, adoption was the first struggle, which subsequently lead to an explosion of different use cases of the Internet – from social networking like MySpace, the birth of search engine companies like Google, and then the entire e-commerce industry.
As suggested by Urban, this is the same type of growth spurt that can come next for AI, so 2018 might be the time for AI to start living up to some of the hype.
Speaking of hype, taking a look at the popular Hype Cycle from Gartner can help us look past much of this ballyhoo, and get a more realistic view of where AI might be at:
This Hype Cycle includes AI as a one of the megatrends of technology that companies should be investigating for the future of their workforces. The cycle can suggest:
- AI trends such as Virtual Assistants, Smart Robots and Conversational AI were reaching a peak of inflated expectations 6 months ago (when the cycle was published).
- Looking forward into 2018, we may see a little more hype, followed by a dip in expectations as companies deal with the realities of implementing these technologies effectively.
This dip in expectations is natural though, as we all want super intelligent robots cooking our dinners, and being friends with us right now! Popular TV show, Black Mirror, even has some of us sold on the idea of living forever in the form of a conscious robot – a bit like Sophia Bot from Hanson Electronics:
Downloading a Michael Jackson bot would be amazing, but we can’t moonwalk before we can walk; like most complex technologies, AI will be implemented bit by bit, and we’ll have to learn from our mistakes along the way.
Take Microsoft’s intelligent chatbot called Tay as a prime example. Tay was released into the wild world of Twitter to learn and communicate with humans but unfortunately was taught to be a controversial racist in less than a day.
Additionally, we may not be quite so ready to trust our bot friends to accurately take our orders without the chance of them mixing things up. For example, when using speech-to-text software to say:
“This new display can recognise speech”
The (not-so) reliable iPhone’s voice recognition can easily interpret that as: “This nudist play can wreck a nice beach”
The Future is Soon
Despite the obvious issues outlined above, PwC still forecast AI to bring economic benefits of more than the current output of China and India combined by 2030. The Hype cycle also predicts it’ll be just 5–10 years until AI reaches a plateau in productivity.
In the grand scheme of things it’s fair to say 2030 is not actually that far off, so preparing your companies and workforces for the robot economy now is a good idea. For when the robots do come for our jobs, the crucial problem will be to create those new jobs that humans can perform better than algorithms (EY – Ernst & Young).
BY GRAEME FULTON