Digital Understanding

Lord Rees of Ludlow Excerpts
Thursday 7th September 2017

(6 years, 8 months ago)

Lords Chamber
Read Full debate Read Hansard Text Read Debate Ministerial Extracts
Lord Rees of Ludlow Portrait Lord Rees of Ludlow (CB)
- Hansard - -

My Lords, I join other speakers in congratulating the noble Baroness, Lady Lane-Fox, on this debate. Our lives have been hugely enriched by consumer electronics and by web-based services that are free or very cheap. Indeed, during a decade where many people’s real wages have fallen, the main reason why they may enjoy greater subjective well-being is the consumer surplus offered by the ever more pervasive digital world. However, it is not an unalloyed piece of good news that young people spend so much time online, and there are other concerns. What about, for instance, the burgeoning information about us on the net—about health records, google searches, where we have travelled and what we buy?

When we are at home, Amazon’s home robot is recording what we say. Even the humble robotic vacuum cleaner can record the floorplans of our rooms. All this information has commercial value to the companies that dominate the sector. Criminal hackers can steal our identity. As the internet of things becomes more pervasive, they will be able to sabotage our house and our car as well. When on the phone or online, it is increasingly hard to tell whether you are dealing with a real person or with a computer. Bots can engage in increasingly sophisticated dialogue—but it is important that we should be able to recognise them for what they are. Would we be happy if a stranger who sat near us on a train could access facial recognition software, identify us and then search our online presence?

AI will enable machines to control traffic flows, the electric grid and such like. They will do such jobs better than humans and that is an unambiguous benefit, but when machines decide the fate of individuals, one is ambivalent. If individuals are denied a request, they should be entitled to be told the reason. One genuine dilemma is that machine learning leads to algorithms that seem reliable, but no human understands how they come to their decisions.

When so much business, including our interaction with Government, is done via websites, we should worry about, for instance, an elderly or disabled person living alone who is expected to access the benefits system online. Think of the anxiety and frustration when something goes wrong. Such people will have peace of mind only if there are enough adequately trained human beings in the system to ensure that they can get help and are not disadvantaged.

This leads to a more general point. The digital revolution generates huge wealth for an elite, but preserving a healthy society will require massive redistribution of wealth and, of course, redeployment of labour to ensure that everyone still has worthwhile employment. To do this we should surely hugely expand the numbers of public service jobs where the human element is crucial and where demand is huge, and now hugely unsatisfied, especially carers for young and old, and in particular, enough computer-savvy carers to help the old and the bewildered.