Artificial intelligence is not eliminating jobs in the software industry – capital is renegotiating their price. This May Day, as the world celebrates labour, the more urgent question is whether technology is being used to empower workers or to make them cheaper, more replaceable, and easier to control.The excitement around artificial intelligence today is neither misplaced nor trivial. However, it is also not unprecedented. If one looks back to the Industrial Revolution, the central question was how to overcome the limits of human muscle. Machines emerged to automate repetitive and physically demanding tasks. Over time, automation became the defining feature of industrial modernity. What is unfolding now is an extension of that trajectory – from muscle to mind.The idea that machines could assist, or even replace, certain forms of human cognition has been around for over a century. By the early twentieth century, computing machines had begun to take shape, outlined by mathematicians, physicists and philosophers deeply engaged with questions of society and politics. In 1950, Alan Turing asked a foundational question: can machines think? His work anticipated many of today’s debates on artificial intelligence. Yet, in today’s marketplace-driven discourse on AI, such intellectual foundations are often quietly sidelined. This is not accidental. It reflects the priorities of capital.The computing revolution that gathered pace in the 1960s and 70s is now at its peak. If the Industrial Revolution took over a century to reshape labour relations – culminating in events like the Haymarket Affair – then the digital revolution is moving at a far faster pace. Artificial intelligence and quantum computing are accelerating this transformation. What remains uncertain is not what these technologies can do, but what they will do to society. This is where a May Day perspective becomes relevant.Recent remarks by influential industry figures reveal the underlying tension. N. R. Narayana Murthy’s suggestion of a 70-hour work week, justified through personal experience, reflects a moralisation of labour that sits uneasily with contemporary realities. Elon Musk goes further, dismissing formal education and privileging demonstrable output over institutional learning. These are not isolated opinions; they reflect a structural shift.Murthy, a programmer himself, understands the difference between writing a simple algorithm and building a real-world software system. While AI tools can generate basic code instantly, they struggle with the evolving complexities of real-life applications – be it a school’s examination system or a company’s financial workflow. Software is not merely code; it is adaptation under constraint. This is precisely why human programmers remain indispensable. And yet, the dominant narrative suggests otherwise – that AI will soon replace programmers. This narrative serves a purpose. It creates uncertainty. And uncertainty, in labour markets, is a powerful instrument to push wages down.We have seen this before. Two decades ago, data entry operators – engaged in repetitive, low-skill work – became easily replaceable, and their wages collapsed. The attempt now is to extend a similar logic to software professionals. In essence, what is at stake is not the disappearance of work, but the restructuring of class.The global technology workforce, particularly in countries like India, has largely been drawn from the middle class. The current shift appears aimed at gradually eroding that position, pushing skilled programmers closer to precarious labour conditions. Artificial intelligence becomes the convenient justification. The irony is striking. Even as companies demand increasingly sophisticated programming talent, they simultaneously propagate the idea that such talent will soon be obsolete. This contradiction lies at the heart of contemporary capitalism.There is also a deeper intellectual vacuum. In earlier eras, thinkers like Albert Einstein or Rabindranath Tagore engaged critically with science, society and ethics. Today, despite the overwhelming influence of technology, such voices are far less audible. Public discourse is increasingly shaped by spectacle rather than reflection. In such a climate, technological change risks becoming detached from social responsibility.Consider a simple example. In classrooms, students are often discouraged from relying on AI tools so that they can internalise problem-solving. Yet in industry, those same tools are widely used – often productively – to accelerate development. The reality lies in recognising that AI is neither a substitute nor a threat; it is a tool that enhances human capability when used critically.However, tools alone do not determine outcomes. Power does. As long as the logic of profit dominates, there will be attempts to use AI not to empower workers, but to discipline them – to extract more work for less pay, to weaken bargaining power, and to normalise insecurity. That is why May Day still matters.The question is not whether AI will transform the world – it will. The question is who will shape that transformation, and in whose interest. In the software industry, as elsewhere, workers must recognise that their value lies not just in the code they produce, but in the knowledge they carry and the systems they sustain. Artificial intelligence, for all its advances, still depends on human intelligence. To forget that is not technological progress. It is a negotiation and not one being conducted on equal terms.Subhamoy Maitra is a Professor of Computer Science at Indian Statistical Institute, Kolkata.