Share

Don’t believe the hype

Principal Advisor, Productivity Commission
17 May 2019

Predicting technology is the first step towards predicting the labour market impacts of technology. But history tells us that technology prediction is like a game of chance. Make enough predictions and you’re sure to get some right… and many wrong, sometimes spectacularly.

The president of IBM, Thomas Watson, purportedly said in 1943 that “I think there is a world market for maybe five computers”. There’s some doubt over whether he actually said it; however, there is little doubt that IT insiders who should have been “in the know” have made some monumental mis-predictions. Here’s a few of my favourites:

"Machines will be capable, within twenty years, of doing any work a man can do." Herbert Simon, 1956

"There is no reason for any individual to have a computer in his home." Ken Olsen, founder of Digital Equipment Corporation, 1977

"Almost all of the many predictions now being made about 1996 hinge on the Internet's continuing exponential growth. But I predict the Internet will soon go spectacularly supernova and in 1996 catastrophically collapse." Robert Metcalfe, founder of 3Com, 1995

"Apple is already dead." Nathan Myhrvold, former Microsoft CTO, 1997

"Two years from now, spam will be solved." Bill Gates, founder of Microsoft, 2004

 "Next Christmas the iPod will be dead, finished, gone, kaput." Sir Alan Sugar, 2005

We might now live in an always-connected world of big data, yet prediction is no less difficult.

Technology won’t develop without optimists. New ideas need research funding and seed capital. And attracting money in noisy, competitive environments requires a relentlessly positive message – downplaying timeframes, costs and difficulties, while overselling latent demand and social benefits.

Technology needs pessimists too. They put a brake on the exuberance of optimists and help kill off truly dumb ideas. They identify social harms overlooked or downplayed by optimists. Pessimists jump enthusiastically on optimistic predictions that fail to eventuate, asking optimists to explain themselves.

The public interplay between optimists and pessimists drives a pattern called the Gartner hype cycle.

Initially optimists drive the cycle, pushing hype towards an early “peak of inflated expectations”. As their overly hopeful predictions fail to eventuate, the pessimists dominate, pushing hype back towards the “trough of disillusionment”. Surviving technologies slowly climb back out.

Peak hype comes a long time – typically years –– before the “plateau of productivity”, when a technology is commercially available and affordable. At this point it starts contributing to increased profitability and productivity, and thus affecting labour markets.

What can we take away from this for the future of work? I can see a few things.

  • A lot of noise about a technology is likely to be a poor measure of its current impact on the labour market. And it may well be a poor predictor of its future impact, particularly as not all hyped technologies survive.
  • The future is unlikely to be dominated by a single “super” technology, rather it will be determined by the interaction of many arriving at different times.
  • It is inherently difficult to answer questions like “is technology accelerating?” and “when will technology disrupt labour markets?” (I’ll have a go at these questions – or at least a framework for thinking about them – in a forthcoming post.)
  • We need both techno-optimists and techno-pessimists. Both are likely to mis-predict the future. Understanding how they interact might bring us a little closer to realism.

Lastly, there is another variety of pessimism. That one grabs the optimistic schedules and market reach of new technologies from the techno-optimists and combines them with social harm concerns – potentially exaggerated – from the techno-pessimists. While such predictions play well to technophobes and others fearful of change, they are no less likely to be spectacularly wrong.

This variety of pessimism offers a business model (actually it’s a pretty old business model) – to make and promote predictions that stoke public fears, and then sell “modelling” and/or “policy advice” on what to do about it to governments. Just another reason not to take predictions of doom at face value.

More information

Share
Tags:

Have your say

Please tick the box to prove you're a human and help us stop spam.

Comments

  • Gravatar for Slothsayer

    Slothsayer 12 Aug 2019, 06:03 (4 months ago)

    And I guess some believed the earth was flat. Predictions are there to be proved wrong or not, and if the future of work requires someone to know the next number after 1 2 4 8 16 32.???... then I might have a way to pay my landlord and buy food We need a Citizens Income if we are too consume and feed the machine. Or maybe a monet-less society, but then unsure how I get fed.

    Please tick the box to prove you're a human and help us stop spam.
  • Gravatar for Greg van Paassen

    Greg van Paassen 12 Aug 2019, 06:02 (4 months ago)

    The histories of actual technological changes can provide a great deal more insight into how future technological changes are likely to proceed than the history of predictions.

    One of the most well-known students of past technological innovations is Vaclav Smil. Pretty much all of his books are worth reading and keeping in mind when thinking about how and when future innovations will change the nature of work. Smil's findings on the time taken for major innovations to bed down can be crudely summarised as "it takes generations [of people], if there is active policy encouragement."

    It's also worth bearing William Gibson's aphorism in mind: "the future is already here -- it's just not very evenly distributed". In other words, the future will be 90% like today, 5% like stuff that you expect, and 5% weird stuff that you never thought of.

    I have two criticisms of the Gartner chart. One is that the "plateau of productivity" is far too high. It should be nearly at the trough, for the chart to be to scale. The other criticism is that, as you mention, Dave, the chart has a huge survivorship bias. Most would-be innovations are as useful as inflatable dart boards, so the "plateau" runs along the x axis nearly all the time.

    Please tick the box to prove you're a human and help us stop spam.
  • Gravatar for Amy

    Amy 20 May 2019, 10:48 (7 months ago)

    Another great post, Dave. Re the pessism that "offers a business model...to make and promote predictions that stoke public fears, and then sell “modelling” and/or “policy advice” on what to do about it to governments" -

    To be fair, I think tech optimists are doing this too. The message is "We can help you seize this transformative opportunity" rather than "We can help protect you from doomsday", but the breathless tone is similar.

    Please tick the box to prove you're a human and help us stop spam.
    • Gravatar for Editor

      Editor 20 May 2019, 14:33 (7 months ago)

      Thanks Amy, I hadn't thought of that. I think you are describing another combination - the plus side of the techno optimist position (significant private and social benefits from new tech) with the delayed schedule of the techno pessimists position. The sales pitch to government also sounds high minded but may be similarly self interested - Dave

RSS feed for comments on this page | RSS feed for all comments