Artificial intelligence (AI) is a huge issue for trade unions. Our movement was forged amidst the transformational change of the late industrial revolution, and we must now get to grips with the convulsions of the tech revolution. Unions know we can’t stop change – but we can influence it, so we minimise the risks and maximise the gains for working people.
AI is already reshaping our workplaces and our economy at pace. It is transforming the education and skills landscape. And it is making decisions which affect millions of workers – about recruitment, performance, promotion, redundancy and dismissal.
That’s why the TUC is leading the calls for effective regulation of AI at work. We want urgent action to ensure people are protected from the risks – and we also want to make sure everyone benefits from the potential opportunities. If we get our approach right, with a strong voice for working people, AI can make work safer, more productive and more rewarding.
But as things stand, trade unions have real concerns about the potential for inequality, unfairness and work intensification. When the TUC polled workers managed by AI, many said the experience was dehumanising, others reported loneliness and constant pressure, and some said they were treated unfairly. As the British Academy’s Hetan Shah has argued: “the worry is not that we’re going to be taken over by robots, but that we’re treated like robots”.
And there’s a real risk of AI-driven gender, race and class discrimination. In October 2021, Uber drivers went on strike about the way its facial recognition software discriminated against BME people, which had led to scores of drivers losing their livelihoods, because they could not log on. Microsoft, which made the software, belatedly conceded it did not work as well for people of colour.
All this underlines the importance of a strong worker voice in shaping AI, counterbalancing the powerful vested interests of big corporations and big tech. The TUC has called for new technology agreements between unions and employers, for proper investment in workforce skills and adult education, and for the gains of new tech to be shared fairly. Above all, we want a legal framework that is fit for purpose.
That’s where the TUC’s AI Bill comes in. Drawn up with lawyers, tech experts and employers, our draft law showcases the rights and obligations we need to protect workers against AI-powered decision making, give them a say over technology at work, and deliver the certainty that business needs. Our research shows there’s overwhelming backing from the British people for stronger protections at work – including a new right for workers to be consulted about the introduction of AI.
Although the King’s Speech promised “appropriate legislation to place requirements on those working to develop the most powerful artificial intelligence models”, the new government has so far said relatively little about how it intends to regulate AI more broadly so there’s a big chance for influence here.
After years of inaction under the previous government, the UK has serious ground to make up. The EU has already passed its groundbreaking AI Act, while there is a growing movement towards context-specific AI law. In the US, Colorado recently passed an AI and Employment Bill, following New York state. The US Labor department published its AI worker wellbeing principles earlier this year. Germany has empowered unions to negotiate about AI through works councils. And Italy is now working on AI at work legislation.
As the voice of working people right across the UK – including academics, IT programmers and systems engineers – unions want to shape the future of workplace AI here. It’s time to put our collective expertise and experience to good use.
That’s why the TUC will continue to press the government to deliver the comprehensive action on AI working people need. In his speech to this year’s TUC Congress, prime minister Keir Starmer promised a new “politics of partnership”. And where better to start than with collaborative working to deliver AI that works for all?