AI and employment: algorithmic control threatens workers’ autonomy and dignity

by John Bratton

• In the last issue, I explained that artificial intelligence (AI) is neither ‘artificial’ nor ‘intelligent’ and algorithms — the lists of instructions that are used to perform certain routines or decision-making — are inscribed with an ideology, choice preferences, and values. AI therefore is not neutral.

In this issue of the Voice, I explore how AI already poses tangible harms to our planet, society, and workers.

AI may invoke ideas of ‘clean’ iCloud architecture, but AI’s carbon footprint is growing exponentially because its ecosystem uses vast amounts of energy, mainly sourced from fossil fuels, which result in substantial GHGE.

In Capital I, Marx makes a pioneering contribution to contemporary debate on ecological destruction, ‘capitalist production… develops only by sapping the original sources of wealth — the soil and the labourer.’

The growth of AI drives demand for ‘rare’ minerals, like white lithium crystal, known as ‘grey gold’, that build computer’s core components. The need for capital to control the supply of rare minerals is behind Trump’s threat to invade or ‘buy’ Greenland, rich in minerals and fossil fuels.

Extracting the rare minerals requires an army of ‘invisible’, low-wage workers in the Global South. Ecological destruction and the exploitation of workers exist at all stages of the AI pipeline.

Theorising about technology, work, and societal change has deep historical roots and centre around six work-related themes — job loss, time, skills, control, surveillance, and the interconnection between work, gender, and family — which can provide a framework for critically interrogating the effects of AI.

‘No job is needed’ — Musk
Predictions that technology causes job losses have featured in debates since the Industrial Revolution. Fast forward 200 years, and pundits predicted that microcomputers (the silicon ‘chip’) would eliminate jobs and be a harbinger to a ‘leisure society’.

Elon Musk’s forecast that AI will do everything and ‘no job is needed’ reignites the debate. A recent forecast predicts that AI could replace 70% of tasks in computer-based jobs. Whereas past technologies eliminated manual work, it’s forecast that AI will replace intellectual and creative work. But we need to take account of workers’ agency when determining AI-job outcomes.

Technology and time are interwoven. Factory machinery intensified work and was enforced by the ‘tyranny of the clock’.

Henry Ford, writing in 1922, explained the importance of time in his automobile plants: ‘The idea is that man [sic]… must have every second necessary but not a single unnecessary second’. Time itself became a key space for workers’ struggle.

Timed routines are no less important in the gig economy. Algorithmic apps empower capital to manage time in terms of what workers are doing, where they do work, and to pay only on demand.

Technology has been paired with the degradation of skills. The thesis, in brief, is that technology deskills workers, reduces their autonomy, and enhances employers’ power over the labour process.

In his analysis of machinery, Marx observes: ‘Along with the tool, the skill of the workman in handling it passes over to the machine’.

The crucial assumption is that technology is not neutral but can reinforce managerial control. AI will change skillsets, some will become more important, while others become less important or even eliminated.

Closely related is the control of work effort. Capital’s instruments of control over the labour process are designed to bridge the gap between workers’ potential and actual level of effort, physical or mental.

Algorithmic control is evident in different domains — Amazon warehouse workers, Uber drivers, and fast-food services — threatening front-line workers’ autonomy and dignity.
‘Digitalised cage’

A key aspect of capitalist work is surveillance. Max Weber coined the well-known metaphor — the ‘iron cage’. — to describe the modern condition. In revisiting Weber’s metaphor, AI-mediated work ensnares workers in a ‘digitalised cage’.

The digitalised cage constitutes invisible surveillance — either in the corporate office or in the workers’ home — making it difficult for workers to challenge. Unsurprisingly, ‘surveillance capitalism’ is a major concern for human rights and trade unions.

Links between work, gender, and family predate capitalism. Men engaged in waged work and women in unpaid domestic labour. Bourgeois ideology assumed men to be the ‘breadwinner’. This ‘normalisation’ of work practices justified women’s’ exclusion, and subordination.

AI-based digital work in the home risks the disolvement of the boundary between working life and other life and underscores the risk of continuing gender precarity and injustice.

We need to draw lessons from the past and the present experience of workers to develop effective strategies to defend working people from AI.

These strategies I explore in the next edition of the Voice.

• John Bratton is co-editor (with Laura Steele) of ‘AI and Work: Transforming Work, Organizations & Society in an Age of Insecurity’ — published by Sage, January 2025

---

Subscribe now to receive the regular Voice newspaper PDF
https://scottishsocialistvoice.wordpress.com/subscriptions/