By David Tuckwell on Thursday 17 August 2017
There is no academic consensus on what AI even is, let alone its job stealing potential
Contrary to headlines and periodicals from businesses, there is no consensus on what AI is, let alone the number of jobs it will steal.
Robo-advice has long been touted as a death knell for financial advisers. As washing machines rid the world of laundry maids so digital wealth management will send financial advisers out the door.
A new study by Opimas, a management consultancy based in London, is the latest piece of research confirming this impression. It found that up to a third of asset managers could be wiped out by advances in technology within ten years.
“Worldwide, by 2025 we expect AI technologies to reduce employees in the capital markets by 230,000 people. The asset management industry will shrink most, with around 90,000 people being replaced by machines,” the report concluded.
Reports with conclusions like this are common, especially from big businesses. Research from McKinsey, JPMorgan, Accenture and Citi Bank have all put out studies stating that the robots are coming. Elon Must at Tesla and Jack Ma of Alibaba have gone so far as warning that robots could threaten civilisation.
But the picture among academics, experts and researchers is a lot more tranquil.
“There is no agreement on what AI is,” Professor Michael Milford, an AI researcher at Queensland University of Technology, told Robo-Advice News.
“The field is very split on what effects it will have. There are a lot of very smart and experienced people who have polar opposite views. There is no consensus.”
But when it comes to the effect robots will have on jobs, there seems to be three camps.
First, there are those who are optimistic about technology but pessimistic about jobs. This camp is headed up by Carl Benedickt Frey, a researcher at Oxford University and director of its centre for technology and employment.
His widely-cited 2013 study claimed that half of the jobs in the US could be computerised by 2023. At particular risk were jobs in transportation and low-end admin staff, which could easily be ground under the wheel of self-driving vehicles and algorithms.
In the second camp are those who are optimistic about tech and jobs. James Besson at Boston University is chief among them. His work found that that banks introducing ATMs increased banking employment, not undermined it.
And tech boosting jobs is the historical norm. Another example he gives is the textile factories in the 1800s, where 98 per cent of the weaving work was taken over by machines. Despite the automation, the number of jobs in textile manufacturing increased over the period.
In the final camp are those who think little of AI and thus little of its prospects to kill jobs. Chief among these is the technology luminary Jaron Lanier, who states bluntly: “there’s no such thing as an intelligent machine. That’s a fantasy that’s used to manipulate people.”
Lanier argues that the whole discussion of AI stealing jobs is based on a fallacious conflation of intelligence and algorithms. Algorithms can be refined and brilliant but they aren’t intelligent. And if any employer decides to replace a worker with an algorithm that’s just that: a decision made by an employer, not the upshot of some economic law.
But where does leave robo advice? In the same place as everything else, Professor Milford says.
“Capitalism is all about earning money. In that context automation is a modern and powerful tool for increasing efficiently or whatever of companies.
“One of the big questions around automation and AI is not about whether it will replace jobs – it almost certainly will replace some jobs – the question is whether it will make new ones. If it doesn’t happen that’s a big issue for society.”
This article first appeared on www.roboadvicenews.com.