Workers and Machines #5

What we’re up to 

At the government’s AI Adoption Summit this month ministers led with a message of ‘pro-worker AI’. We’ve been pushing hard behind the scenes to bring a much-needed shift that centres workers. The messages are starting to go in the right direction – Now we need robust action. 

The TUC’s assistant general secretary Kate Bell talked AI on the Today Debate. Is there a crisis of control in AI? Have a listen here if you missed it. 

At UNISON’s delegate conference, Adam spoke at an event on Big Tech in public services alongside legal campaigners Foxglove and health justice organisation MedAct. 

Movement movements 

AI Early Careers Alliance new AI and Economics Institute: At the AI Adoption Summit The government has announced new initiatives, to monitor and address how AI is impacting early careers as well as a dedicated body to assess the current and potential impacts of AI on the labour market the wider economy. Mike Clancy, the general secretary of Prospect union will be involved in both. 

NHS tech workers want to cut ties with Palantir: NHS Analysts Together, a self-organised group of data and digital workers across the NHS, have launched an open letter asking the NHS to cut its contract with Palantir – the same recommendation the Science, Innovation and Tech committee recently made in its first report of the 2026-27 session, Rewiring the state: Delivering digital government. The letter is accepting signatures up to 21 June. 

Meta has scaled back its employee tracking (sort of): After pushback against its plans to collect employee data to train AI agents, Meta has now said employees can pause data collection for "up to 30 minutes at a time" and request exemptions from the initiative altogether. Workers seem less than impressed, as one told the BBC it felt "very dystopian". 

Two unions are taking on the New York Times over AI surveillance and monitoring: Tech workers at the Times have said their collective bargaining agreement was violated by introduction of two internal AI tools that track and evaluate employee performance and activity. The Tech Guild (tech workers) and the Times Guild (editorial and business workers) have both taken legal action against the Times, after the publication refused to respond to their requests for information on AI use. 

Elsewhere in America, platform drivers have won union recognition: The App Driver's Union, comprised of Uber and Lyft drivers in Massachusetts, was officially recognised at the end of May and is the first of its kind in the United States. While federal law only allows traditional employees to form unions, state-level legislation enabling drivers to unionise appears to be having a domino effect. A ballot was passed in Massachusetts in 2024 and California’s governor signed in legislation in 2025. A similar bill in Illinois is currently making its way to the Governor's desk to hopefully be signed into law.  

Samsung workers in Korea win a piece of the (chip) pie: Workers at Samsung Electronics in South Korea won a landmark profit-sharing agreement after a months-long dispute over the company’s ballooning AI profits. The agreement, which was mediated by the government, averted a strike that was set to disrupt the global supply of chips and cut 0.5 percentage points off Korea's economic growth this year. The company’s 78,000 semiconductor workers will be given 10.5 per cent of the company’s operating profits. Workers in other, less profitable divisions like consumer electronics aren’t jazzed about the deal. 

Bookmark this:   

Check out this model language for negotiating on AI from The American Federation of Government Employees (AFGE).  

In brief: AI driving more bad work 

AI is letting us do more, faster. We've been promised scientific breakthroughs, expanding access to information, and leisure time in the place of admin tasks. At this point though, many workers are just getting an ever-increasing volume of content of very questionable quality. 

This kind of ‘workslop’ usually comes from colleagues. Sometimes it is hidden usage, other times it's because management have set blunt targets for AI usage. Worker’s voices in defining whether and how AI should be used at work is key to avoiding this – a fact that is slowly being recognised. 

But now it’s coming from service users, creating new barriers for workers to accomplish the task at hand, especially for those whose work relies heavily on professional judgement. 

Doctors and nurses are seeing patients who come with a diagnosis from ChatGPT. The Information Commissioners Office has had to issue new guidance for public authorities on how to deal with AI generated FOI requests, which practitioners say are longer, more complex, and less clear about what information is being requested. A recent study of human resources professionals found that 84% of HR teams report heavier workloads as AI-tailored applications increase.   

Unions are experiencing these same challenges. Reps and officers are fielding huge volumes of AI assisted casework from their members, and much of it is flawed. Often, it has misunderstood employment law -- or has defaulted to the American context. It doesn't ask complex questions, instead taking what users say at face value, so it misses the subtext and nuance that a union rep might flesh out. Frequently, what AI will suggest is an escalation – jumping to taking legal action when a negotiation would prove more effective.  

In many ways, it’s understandable why people turn to these tools. They want to know how to best tackle challenges at work or support with a job application. They want to reduce friction between themselves and their desired outcome. Instead, too often they are inserting a layer of pseudo-expertise for someone else to untangle.  

Events etc 

At a Crossroads: Worker Voice and the Future of Workplace Technology Workplace technology is not inevitable. It is shaped by choices. From design and development to deployment and governance, those choices affect job quality, worker wellbeing, autonomy, productivity and organizational performance. Join this ILO webinar to explore how workers and their representatives can have a greater voice in shaping the technologies that affect their work.  

 June 23 | 14:30–16:00 BST | Online   

Campaign Against Climate Change - This Is Not Inevitable: Standing up to AI and Big Tech. Session 3: How AI is affecting our jobs and wider society, and the action that trade unions are taking 

Speakers: Simeon Gill, President of University of Manchester UCU; Laurence Bouvard, Equity, Screen and New Media Committee member; Mansoor Ahmed, Unite EC, Passenger Transport, United Tech and Allied Workers speaker (invited)  

 June 27 |10.30-16.30 | Mechanics Institute, Manchester