TThe real danger that artificial intelligence poses to jobs goes beyond just losing your job. It’s a growing gulf between those who use AI to develop their skills and those whose working lives are increasingly shaped by opaque AI-powered systems of surveillance and control.
The debate about artificial intelligence and how it will impact workers is stuck in the wrong place. But there are also warnings that machines will replace millions of jobs. The other is the claim that AI will significantly improve productivity. Both stories miss what is already happening in workplaces around the world, from the UK to Kenya to the US.
For some, AI can help remove the monotony from their daily tasks. These are often people in higher paid, more autonomous roles such as analysts, consultants, lawyers, academics and managers. In these jobs, AI can feel like co-pilot if it is deployed to augment employees rather than replace them. It can support human judgment, speed up daily tasks, and create space for more creative thinking.
But for many others, AI is not an assistant. This is my boss.
This is displayed in scheduling and monitoring tools, route optimization software, and automated performance dashboards. All these systems determine who gets what shifts, how long tasks take, and whether someone is performing to their fullest potential. These workplaces are not meant to use AI. It is what watches over you and controls you.
That’s the new divide we all need to focus on.
A third of UK employers already use “bossware” technology to monitor the online activities of their workers. This already widespread worker surveillance offers a glimpse of what’s to come.
This is why the question of whether AI is “good” or “bad” is pointlessly broad. The truth is even more nuanced. Employers are using AI to empower some workers and subject others to more intensive and inhumane supervision. It is creating new opportunities at the top of the labor market while tightening control at the bottom.
Furthermore, in the future, the same algorithmic control and monitoring techniques honed in warehouses, delivery vans, and gig work platforms could find their way into corporate headquarters, hospitals, and schools. We’re already seeing this phenomenon at companies including Amazon, whose software engineers say they are being monitored and pressured to use AI to increase productivity, even when it counterintuitively slows them down. Meta also plans to track and capture employee keystrokes, mouse movements, and clicks to train AI models. Some of the same workers who are currently benefiting from the rise of AI are ultimately poised to lose that advantage.
My own research over the past decade on the coexistence of workers and AI, cited in the 2024 White House Economic Report, suggests that the most pressing issue regarding the impact of AI on jobs is not immediate mass unemployment. It’s a growing gap in skills, autonomy, and well-being between those who have come to work with AI and those who feel controlled by it. Many jobs will remain in the future, but they will be more pressured, more fragmented, and less human.
This is important because work is not just about making money. It’s also about dignity, trust, and control.
During the pandemic, many people became acutely aware of how much impact work can have on their mental health. A workplace managed by AI will only increase work pressure. When every click, step, call, and pause a worker makes is measured and evaluated by a system that cannot be fully verified or challenged, the impact is stressful.
For those in warehousing, retail, hospitality, logistics, customer service, and the gig economy, that could mean being driven more strongly by systems that pretend to be neutral, objective, or efficient, even though they are never.
This is not just a technical issue. It is social, political and moral.
Take the UK, which wants to appear ambitious about AI. There are now big plans to expand AI skills across the workforce. It all sounds positive. But behind this rhetoric lies a more unpleasant reality. Many organizations are still ill-prepared to deploy AI equitably.
A recent global survey of business leaders found that while most companies say AI skills are a source of competitive advantage, relatively few are investing meaningful budgets into developing AI skills for their employees. Even fewer companies have strong governance in place. Many managers still have little actual responsibility for helping their teams adapt. That’s how inequality is reinforced.
If higher-wage workers are trained to use AI and lower-wage workers are exposed to it simply through surveillance and automated controls, this is not a story of shared progress. This is a story of deepening imbalance.
Workers across the economy need access to meaningful training that not only helps them use digital tools, but also builds a wide range of skills that will become even more important in the AI era, such as judgment, communication, and critical thinking.
Basic democratic principles are also needed in the workplace. Systems that affect pay and performance must be transparent and challengeable. Above all, workers should have a say in how these technologies are implemented. AI should not be used by people behind closed doors and justified in terms of efficiency. It should be shaped by the people whose lives it impacts, and research shows that involving workers in the process improves the quality of work and helps employers integrate AI more effectively.
Choices about how AI reshapes work are not being made in Silicon Valley boardrooms or summit halls. It is now being made in workplaces across the UK and around the world. And unless we pay attention, new inequalities driven by AI will arrive quietly, becoming deeply embedded and only noticeable once they are already widespread.
