Since the rise of the industrial revolutions, there has always been a very active debate among the academic community confronting “technology innovation optimist” and “technology innovation pessimist”. Likewise, there is debate whether artificial intelligence (AI), like many previous technologies could have a positive or a negative impact on the economy, especially on jobs.
How to approach the impact of innovation on the economy?
According to economic literature, there are several distinctions to make when it comes to impact of innovation on the economy:
First, there is a distinction between the micro, industry and macro approach. Each individual business is actively focused on productivity and human resources savings, but the job reduction resulting of it could be compensated at the industry and country level through price decrease and new market demand.
A second distinction should be done between process innovation and product innovation. A process innovation increases productivity and destroys jobs, a product or service innovation open new markets and therefore create new revenues and jobs.
So, innovation has a double impact on the economy, on one hand job destruction and the other hand job creation. The national statistics tend to measure job destruction rather than job creation, fueling the “technology-pessimist” approach.
AI increasing productivity and creating new jobs
There are definitely areas where AI has potential to improve productivity. In manufacturing, for example, there are benefits from intelligent automation and use of the Internet of Things and AI. The immediate benefit is improving the maintenance of costly production systems by identifying and shortening the time to operate it.
In fact, economic research shows that the job destruction debate is the wrong debate. More important with technology innovation is that the structure and the type of jobs is changing deeply. Take the digital economy; new jobs have been created like digital manager, digital designer, digital project manager and so on. Similarly, AI creates a full range of new jobs like Machine Learning expert, Image recognition expert, Cognitive computing expert, AI system developer.
How can you even begin to predict the productivity of jobs that do not exist yet? In the meantime, machine learning has been offering us augmented knowledge for a while now.
Innovation is a question of data. That is no secret. But how does a company transform its data into business models to drive change? That's what we asked 100 business leaders, and we got astonishing insights: Innovation - From data to the business.
What will be the impact of AI on the economy?
According to McKinsey Global Institute’s “Artificial Intelligence: the next digital frontier” research paper, between almost zero and one third of work activities could be displaced by 2030 across 46 countries. But at the same time also 75 million to 375 million workers (3 to 14 percent of the global workforce) will need to switch occupational categories.
There are for sure huge productivity gains with AI, but they have been concentrated in the hands of a few ‘super-companies’, such as Google, Facebook and Amazon, which have been able to take advantage of technology and now dominate their sector. If this is the case, however, there should be some ‘spillage’ down to other companies, and again, it certainly is not the whole answer.Revolution takes time, but I expect #AI to deliver on its potential @MarcelLemahieu #HBRfavourites Click To Tweet
These things take time…
As AI systems start to be more widely used, however, we are not seeing a huge growth in productivity yet nor in market creation.
It is rooted in history, and particularly previous technological revolutions, and suggests that any ‘general’ technology, like AI, needs a wide range of supporting developments before it delivers its full potential. A steam engine, for example, was interesting, but not actually transformative until it was harnessed to wheels, in the form of trains, or to drive other machinery. Electricity took 20 or 30 years to spread throughout industry, because manufacturers were not sure how to reorganize production lines to get maximum benefit.
In other words, there is a lag between the introduction of a general purpose technology, and the full productivity gains from it. Revolution takes time, and we have not yet reached the full transformation promised by the Digital Revolution. Looking at discussions about Industry 4.0, it is clear that even if there are isolated pockets of good practice, we are still a long way from embedding analytics routinely into every aspect of industrial life. This also suggests that although the digital revolution has come a long way, there is still a long way to go.
... but will accelerate once they take hold
Accenture has suggested that labour market productivity could increase by up to 40% as a result of AI, even if there is little sign of that growth happening yet. History tells us, however, that once productivity starts to grow rapidly, as a result of a particular technology ‘coming of age’, future growth accelerates. I expect AI to deliver on its potential, like so many technologies in the past.
We also conducted a SASChat - Innovation@scale on Nov 14. See what participants were talking about. We asked five questions. Find an excerpt of that amazing discussion on twitter:
How is analytics changing the scope of #innovation?
(1) As #innovation is all about changing the way of doing things, Analytics can bring a new dimension of being able to predict what results your innovation efforts will bring (Andreas Kitsios).
(2) I think #analytics is building the scope of #innovation more than changing it. (GorkemSevik)
What is the role of scalable analytics capability in driving #innovation at scale?
(1) #Analytics at scale can support decision-makers in finding hidden patterns in tons of data. It wouldn't be possible in any other way (Federica Ballerini).
(2) To enable your business to have exponential growth ability (Igor Dsiaduki).
How has analytics helped athletes collaborate more effectively with their support teams?
(1) Support teams want to know how an athlete's body and mind are progressing, responding to training and competition and recovering. Analytics may not provide the answer but it allows support staff to start the conversation needed for collaboration and build trust (Reece Clifford).
(2) Absolutely. Things like sleep can be monitored if all parties happy, so that what's happening elsewhere (away from training venue) can help create a clearer picture of the athlete's state of mind and body. After all, the 2 are interlinked! (David Smith).