AI is transforming our world – should we be worried?
In opinion / By Sandra Vogel / 09 November 2016
I recently listened to a news item in the business segment of Radio 4’s Today Programme explaining how financial traders are having difficulty making a buck these days because of the prevalence of computer based trading.
Apparently, computerised trading systems’ lack of intuition makes it difficult for humans because we find it hard to second-guess what they’ll do next.
Where a human trader might buy a stock because it’s fallen to a level they intuitively know is a low, computerised systems make decisions based on patterns without applying judgement.
That makes it tricky for people to work out what a computerised system might do next, and hard to decide whether to buy, buy, buy or sell, sell, sell.
Did I ask you to climb a mountain?
Anyone who works in an environment that has an IT team which controls an organisation’s ‘computery’ will have spent time wondering when their login problem will get fixed, or when they’ll get the promised higher level security clearance that goes with a new role, or why the IT person says ‘no’ to a perfectly simple request to …. (fill in your own blanks…..).
Some of us like to blame the ‘jobsworths’ in IT for this situation, calling out people who seem to enjoy the power they have over what is, after all, the fundamental tool of the trade for most of us these days.
Such people are barriers. They stop us doing our jobs efficiently, make us look for ways round problems they refuse to deal with, or tell us we are in a queue and must wait out turn despite the fact that we’ve got a deadline to meet or a deal to clinch.
If only the whole business were automated, we’d get what we want so much more easily.
A robot IT technician wouldn’t answer back or grind their teeth at us as if we were asking them to scale Kilimanjaro by lunchtime. Instead they’d maybe blink their robotic eyes and trundle off like a compliant Wall-E to do our bidding.
Before they’ve made it back to base the robot will have hooked into the network and fixed the problem.
No “I’m Sorry Dave, I’m afraid I can’t do that” here.
Well, maybe. But it’s not all that likely.
Lather, rinse, repeat
AI does have a big role in the workplace of the future, but this is less likely to be based on supplying us with humanoid gofers than it is to focus on what it is already proving to be very good at.
AI’s success is built broadly around three key functions: automating repetitive tasks, streamlining workflows, and providing access to information.
AI can be used to look at mountains of data, process that data, and draw conclusions from it much faster than humans can ever hope to do. There are many, many ways in which companies can use this.
I’ve already mentioned share trading.
How about recruitment? Automated systems can look at CVs and rule out those a company would not consider, leaving just the good fits for humans to take a look at.
Interestingly, some studies show that this is a good way to remove bias – and analogies have been made with the way ‘blind auditions’ helped transform the male bias in orchestras in the 1970s and 1980s.
When it comes to tech support a virtual agent can deal with all the initial help requests. Password login problems might be designated a high priority and sent straight to a real person for immediate help.
Requests about how to do something can be answered by a bot, without a person ever needing to get involved. Responses come in the blinking of an eye because the AI system is fast and knows where to find the information it needs to provide.
More complex questions which the system can’t deal with can be triaged out to real people.
This kind of virtual agent processing is also employed in customer facing organisations to deal with customer queries or handle customer care.
After triage, real people get to deal with the questions the bots can’t handle.
Wall-E vs HAL
It is a given that AI will play an increasingly large role in business management and financial systems (as well as in retail, healthcare, city management and other complex areas). The rise of the Internet of Things, mountains of data to be crunched through, and the increasing pace of life are going that way like a train running down the tracks.
Little did Oscar Wild know when he coined that phrase ‘life imitates art’ that mobile phones would develop in the US to look like Star Trek’s communicators, or that we’d be building tiny medical computers that can roam around the human body like the micro-ship in the 1966 movie Fantastic Voyage.
Nor could Wild have foreseen how good military drones are at killing people.
This is our dilemma.
What we need to avoid is life imitating The Terminator, we need to make our use of AI more Wall-E than HAL.
On this subject the likes of Professor Stephen Hawking, Steve Wozniak and a myriad of academics and business people are on the same page.
They are among more than 8,500 people to have signed an open letter exhorting the pursuit of social benefits from AI.
Meanwhile, Hawking has gone even further.
In a 2014 editorial for the Independent newspaper, Hawking wrote “One can imagine such technology outsmarting financial markets, out-inventing human researchers, out-manipulating human leaders, and developing weapons we cannot even understand. "
Hawking went on to say "Whereas the short-term impact of AI depends on who controls it, the long-term impact depends on whether it can be controlled at all.”