What is AI and what can we expect to see from it in the future?
You’ve probably heard of artificial intelligence, or AI. Whether you’re constantly keeping up to date with AI news on Wired, or all you know is there are robots involved, you’ve heard about it.
There's ben a lot of talk about the job stealing robots and the risks of AI, but are these fears justified? And what actually is AI?
A 2018 study conducted by Capgemini on AI’s potential in retail showed that retailers could save over $340 billion a year by 2020 as a direct result of AI cross functions, with an expected 80% of the savings coming from increased efficiency and streamlined processes.
With more companies adopting AI software and increasing evidence that it'll be around for a while, we thought we’d investigate its potential.
What is AI?
When most of us think about artificial intelligence and the future, the first thing we think of is robotics. And if you google ‘AI robots’ the first thing you’ll see is a somewhat haunting image of ‘Sophia’. So the misconceptions around AI aren't shocking.
But it's more than just robotics.
Amazon described artificial intelligence as:
“The field of computer science dedicated to solving cognitive problems commonly associated with human intelligence, such as learning, problem-solving, and pattern recognition. Artificial Intelligence, often abbreviated as ‘AI’, may connote robotics or futuristic scenes, AI goes well beyond the automatons of science fiction, into the non-fiction of modern-day advanced computer science.”
Let’s simplify this.
The founders of AI, John McCarthy and Marvin Minsky, defined artificial intelligence as any task conducted by a machine or program that a human would require intelligence to do.
This makes it hard to confidently know what is and isn't AI. McCarthy and Minsky’s definition is pretty broad and leaves a lot of room for debate.
ZD Net technology writer Nick Health breaks up AI terminology into three simple categories:
Narrow AI: Narrow AI are “intelligent systems that have been taught or learned how to carry out specific tasks without being explicitly programmed how to do so.” Health uses examples like Siri and the vision0recognition system on a driverless car. Basically, Narrow AI can't learn anything beyond what we teach.
General AI: General AI is the type of artificial intelligence that draws in the most attention, because, much like a human, it can evolve intellectually and learn on the job. This is where most people see the future of AI going.
Machine learning: Machine learning is the process “where a computer system is fed large amounts of data, which it then uses to learn how to carry out a specific task, such as understanding speech or captioning a photograph.”
Does AI have a place in the future of retail?
The Capgemini survey predicted that the use of AI in retail would lead to a rise in efficiency and customer satisfaction followed by increased revenue
It also revealed a “boost in promotion efficiency by enabling automated ad buying, personalising customers' online experiences and serving users targeted recommendations.”
This seems like a promising future. But how are businesses using artificial intelligence software today?
Amazon was a pioneer in AI as an early user of the technology. They began employing artificial intelligence software as early as the 2000s and saw a 35% increase in their sales as a result.
Using robots to transport stock in their warehouses has made the process twice as fast. It's also reduced task risk and inaccuracy.
The company also uses AI for drone deliveries with Amazon Prime Drone and cashless stores through the use of biometrics and the Amazon Go app.
Retail giant H&M has started using AI to analyse their stock, creating software that can analyse store returns, receipts, and “loyalty cards to predict future demand for apparel and accessories and manage inventory.”
Even the UK supermarket chain Morrisons has been getting in on the action by partnering with AI solutions company BlueYonder. Together they’ve created a solution that replenishes their stock throughout their 491 stores via optimisation and forecasting. This led to a 30% reduction in shelf gaps across all their stores.
While they may not seem groundbreaking anymore, chatbots are AI, and an estimated 80% of brands are either using or thinking about using chatbots in the near future. The implementation of a chatbot on a company site may seem small, but the effects can be massive.
Businesses across the globe spend up to $1.3 trillion a year responding to 265 billion customer service requests. Chatbots help reduce those costs by at least 30%.
Instead of replacing jobs, they’re helping humans do their jobs better. Online bots allow customer support agents, much like our own at SumUp, to focus on more complex requests from customers. Chatbots automate easier support tasks, allowing human agents to dedicate their time to what's important.
It’s clear AI minimises the time we spend on trivial tasks, but is it stealing our jobs?
Why you don’t need to worry about a robot stealing your job
PwC Ireland reported that approximately a third of ‘existing jobs’ could succumb to automation by 2030. While this figure may seem scary, there's more to it.
AI is expected to change jobs by taking over menial tasks and creating new employment. The way we work will change, but humans will be focusing on the more challenging, interesting sides of their professions while AI takes on the tedious work.
Maintaining that that AI is more of a sidekick for his photography business than a competitor, Berlin-based EyeEm CTO and co-founder Ramzi Rizk explains:
“It might seem like these technologies are replacing human creators and consumers, but we believe they’re empowering us by offering creative freedom. Machines won’t take over the work of photographers, they will simply make it quicker and easier. The boring stuff goes out the window and creators have more time to focus on what matters: creation.”
Forbes technology writer Joe McKendrick reflected on a paper written by MIT’s Erik Brynjolfsson in his recent article “Artificial Intelligence Will Replace Tasks, Not Jobs.”
In his paper, Brynjolfsson explains how "machine learning technology can transform many jobs in the economy, but full automation will be less significant than the re-engineering of processes and the reorganisation of tasks."
Brynjolfsson goes on to say that businesses should focus on reducing jobs that AI will take over, and place their employees in positions where they can “spend more time with higher-level tasks.”
So, what jobs does the MIT professor actually think are most likely to face automation? Credit authorisers, morticians, and undertakers ranked highest on his list, with animal scientists, archaeologists, and massage therapists appearing to be the least likely.
There’s still work to be done
Wired writer Tom Simonite sat down with Google researcher and Stanford University professor Fei-Fei Li at the WIRED25 Summit in San Francisco. Li described AI as “powerful” but still a “nascent technology".
At present, AI is an industry that lacks diversity, which means that as the technology evolves, it will take on the social hierarchies and unintentional prejudices of its creators.
Computers are only as good as the foundations we lay for them, and there have been times where algorithms haven’t “captured the complexity” of very human issues. If we rely on technology to handle human tasks such as housing, that technology needs to be as unbiased as possible.
Li has been instrumental in the adaption of autonomous vehicles and facial recognition software, but at the summit, Li argued that “the revolution is incomplete.”
In her opinion, AI needs to focus on building software that works alongside rather than replaces human intelligence.
She estimates it will take 25 years for AI to be fully integrated into our daily lives. So while there is still time to fulfil the necessary changes, “we have to act now.”