The public conversation about new technology always lags far behind the development of the technology itself. We can see it in the sudden moral panic about the impact of social media on our politics and in the nascent public conversation about racist algorithms, both established phenomena that we are only just getting to grips with. So it is that we are yet to have a serious discussion about how AI and other new technology will shape, and is shaping, our experience of work.
The conversation about tech and work is going on all the time, but in a disjointed and incomplete way. There is the “gig economy” conversation, which is too often simply a dialogue about the exploitation that characterises some prominent gig jobs. Then there is the “robots will take our jobs” conversation, with its dire warnings about mass unemployment but which rarely gets beyond arguments about the headline numbers. And then there is the AI ethics conversation, which is totally siloed off from the discourse around work.
Neither the EU Commission’s High-level Expert Group on AI, nor the UK’s Centre for Data Ethics and Innovation, has a remit to look at the impact of AI on workers. TechUK’s latest publication on ‘making digital ethics relevant to the lives people lead’ sets out 8 principles for action. Not one of them mentions workers. If we are not careful, the rules will be set, and workers will have been left behind.
Workers need to be skilled. They need to understand privacy rules and act ethically in the handling and deployment of tech. But largely, they do not have a voice in the industrial and technological changes that will impact them. Ignoring workers in that debate is the sure-fire way to create an AI divide between winners and losers and to incubate a culture of distrust.
All of these strands must be pulled together by anyone serious about understanding and shaping the future of work. Trade unions are one part of that picture, and for trade unions in particular, the place of workers in the new emerging AI economy must be our meat and drink if our institutions, born to address the challenges of the nineteenth century, are to survive and thrive in the twenty first.
The Women Leading in AI group recently launched a set of recommendations to help prevent discriminatory practice in the deployment of algorithms. This is a welcome contribution to a debate that needs to become much more prominent. As the co-founder of the group, Ivana Bartoletti has said: “We want to mobilise politics to get a grip and set the rules around artificial intelligence so that it does not discriminate against any minority and is led by our shared human values.”
AI and digital is the platform for the next state of capitalist development. Today’s moral issues are about surveillance, the deployment of data to manage people, and the impact of human bias on tech development. If we want to get the new rules right then workers need a voice in the discussion.
There are signs that things are changing. Uni Global Union, working with unions like Prospect, has set out a number of practical guidelines on worker’s data rights and privacy and on ethical AI. But there is much more to do.
So, where does this take us?
There are a number of principles that should guide how we have this conversation. First is tone. Recent survey data from Prospect demonstrates that younger workers are more optimistic than older generations about the impact of new technology on their work. Unions and others must be careful to mix warnings about the potential pitfalls of AI and other technologies with ideas for how they can be harnessed to improve working life. If we do not then we will fail to bring along those we seek to speak to.
Second, the conversation about AI ethics must have at its heart the workforce who are designing this technology. The burgeoning consciousness of the tech workforce, whether seen in the recent Google walkouts or the whistleblowing in the Cambridge Analytica scandal, is something that must be welcomed and built on. In fact, one of the greatest arguments for unionisation in the tech sector is to give voice to the ethical concerns of a workforce increasingly alive to the impact of their work on the lives of their fellow citizens. If we conceptualise technology as something that is only done to workers, and fail to address the fact that it is also done by workers, then our solutions will be incomplete and ineffective.
Third, our discussion must also involve the workforce who will be affected or managed via the application of data and new technology. Trust is important in creating an inclusive economy. So too is ensuring “shop floor” checks and balances. One of the most interesting comments at the TechUK summit last year came from a tech lawyer who described AI ethics as something her clients now regard like health and safety. She meant as something that must be taken seriously, rather than being optional. It is a useful way of looking at things. The evidence is clear that the best records for health and safety come in businesses that respect and encourage an independent voice for workers, and involve unions.
Politics and society cannot keep playing catch-up as technology races ahead, transforming the world of work as it is does so. Prospect is working with Uni Global Union to test and develop new approaches to give unions and young workers improved job quality and a voice in the digital world of work. We want to use tech to enable workers to access their own data and to help inform how this economy is developed.
Ultimately, this is a shared challenge. Unions, politicians, businesses, workers and civil society have to shape the consensus around the ethical application of AI and other technology at work or they will be rendered irrelevant to a new generation of workers.
Andrew Pakes is director of communications and research for the trade union Prospect