Richard Exell is a senior policy adviser at the TUC, the Trades Union Congress in the UK, the body that brings together the country’s main trade unions. He is the TUC’s leading specialist on robotics and AI.
How is the TUC/ trade union movement preparing for greater use of robotics/AI in the workplace?
The British trade union movement is really only beginning to think about robotics and AI. We took this very seriously indeed in the 1980s and there is something of a widespread feeling that we were a bit too quick to see a problem there. Because we did not have the crisis of leisure that some people predicted, there is now a feeling of ‘once bitten, twice shy’ in some unions.
But, at the moment, there is a big division between what economists are saying and what futurologists, technologists and sociologists are saying. The economists say that technology does destroy jobs but that it increases overall prosperity and that, very often, that process has been very positive. So the consensus is, among economists, that it won’t be different this time.
But I am inclined to think that there is a strong argument to say that this time it is different. I’ve been talking to researchers in trade unions in the last few months and there are several of us who think that we ought to be doing more.
We’ve got less power to negotiate over the introduction of technology in the way that we used to. Our ability to influence political debate is less than it was.
One important paper in this area is Richard Freeman’s, from Harvard, called ‘Who Owns the Robots Rules the World’. His suggestion is that we need legislation to spread the ownership of robots to the whole of society or to workers in the companies involved. It’s hard to see this being readily accepted. A lot depends on persuading the political world that we need contingency planning for this. Young people might have strong feelings in this area but demographics don’t favour the political influence of the young now.
How do you see the position of the self-employed developing?
One possibility is that the new technology may well change the nature of work. Uber and Amazon’s Mechanical Turk are examples. One possibility is that we could be evolving in the direction of a labour market in which it’s much harder to claim what you are doing is work and, therefore, to claim rights as a worker. It could become much harder to identify who your employer is. On Mechanical Turk you could be doing a series of clicks that are 20 per cent for one employer, 5 per cent for another and so on. You cannot say that the platform you are doing the clicks on is your employer. Asserting rights against someone in these circumstances is going to be tremendously difficult. There is a role here for unions to establish with government that this work needs to be regulated.
We don’t want to see the self-employed doing damage to their rights at work. Uber is a good example. You get delisted as an Uber driver if your car isn’t new enough - but how do you get to pay for the car if Uber can’t guarantee you an income stream?
There is a process going on here: all the risks of the employment relationship are being devolved to the individual worker. There is a real risk that AI and Big Data are further speeding up that process.
We need more awareness of the conditions that people are working in. When people are turning their hobbies into their own businesses, that is a good thing. We don’t want to get in the way of that. But we don’t want them to be exploited because they love what they are doing. There have got to be ways of extending social welfare to cover this.
Where do you think we might end up?
I’ve been very much influenced by a paper from the Oxford Martin School which suggests that there could be two phases - the first affecting jobs in manufacturing and the second in which even the creative and intelligent jobs will start to be hit. Eventually, sometime this century, machines are going to develop creative abilities and, possibly, social intelligence as well. Once that happens, every single job will look shaky. The questions of who owns the robots does then become absolutely vital.
Do we have immense riches for a handful of people who own the capital and everyone else is dependent on what those people are willing to allow them? That small group will have to allow something to the others because, otherwise, there will be no demand for products and services.
Unless we tackle the ownership question of robots, the future is likely to be one of deepening inequality.
.jpg)
.jpg)
.jpg)
.jpg)
.jpg)
