Guardian: AI Is Already Cutting Jobs, and Energy Crisis Could Accelerate Layoffs

Artificial intelligence may hit the labor market significantly faster than previous technological waves, and the new energy crisis only amplifies this risk. When economic growth slows, raw materials and energy become more expensive, and companies gain increasingly accessible automation tools, so the choice in favor of machines becomes almost mechanical for business. In this scenario, the problem is no longer reducible to a dispute about the distant future: the question is how quickly governments will manage to mitigate the consequences for people whose tasks AI is already capable of taking on now.

In his column, Larry Elliott describes this moment through the old idea of 'creative destruction.' Capitalism is constantly being renewed, displacing outdated work methods with new ones, and each such transition is painful for part of the workforce. The difference is that previously, technologies mostly automated physical labor, while the current wave of AI is entering the territory of cognitive tasks. It is no longer just about assembly lines, warehouses, or cashiers, but also about document analysis, customer support, text preparation, basic development, accounting operations, and other office work that until recently was considered relatively protected from automation.

Typically, such transitions proceed more smoothly if the economy is growing and the labor market is capable of quickly absorbing people who have lost their previous positions. Then governments have time for retraining, companies have time for gradual process restructuring, and employees have a chance to move into new roles without a prolonged downturn. However, in the author's view, the current backdrop is almost opposite. Even before the latest round of Middle East conflict, global growth looked weak, and hiring prospects appeared unstable.

After the spike in geopolitical tension, this was compounded by expensive energy, raw material supply disruptions, and more bleak economic expectations. The IMF has already lowered its growth forecasts, which means it is becoming even harder for businesses to maintain staffing levels and even easier to justify accelerated automation. This is where the energy crisis and AI begin to act as mutual amplifiers. When companies face rising costs of electricity, logistics, and materials, they begin to cut costs more aggressively. If at this point there are technologies on the market capable of replacing some office work faster and cheaper than before, management has a strong incentive to remove people from processes.

Optimists respond that history has frightened society many times with 'machines that will take away jobs,' but in the end, new technologies created more employment than they destroyed. Elliott believes that this time there are at least two reasons not to be reassured: AI may prove to be a far more universal technology than previous waves of automation, and new jobs, even if they appear, are in no way obligated to be as well-paid as the ones that disappear.

From this follows a scenario that particularly concerns the author. If automation begins primarily to eliminate well-paid white-collar positions, it will strike at consumer demand. Machines can indeed work around the clock, not take vacations, and not get sick, but they don't buy cars, don't rent homes, don't spend on restaurants and personal services.

The research firm Citrini, cited in the column, described the possible 2028 crisis precisely this way: companies mass-implement AI for efficiency, lay off people, demand in the economy drops, revenues plummet, and business responds with a new round of cuts and further automation. Such a vicious circle is capable of hitting not only employment but also the stock market. The paradox is that the crisis then would not arise from the failure of AI, but from the overly successful implementation of AI at the level of individual companies.

The main conclusion in this logic is eminently practical: it is no longer enough for governments to simply welcome innovation and hope the market will sort it all out on its own. Fast and large-scale measures are needed—retraining, new industrial policy, and redistribution of productivity gains. Otherwise, most of the gains will go to a narrow circle of tech companies and investors, and society will face prolonged pressure on employment, weaker demand, and an economy in which automation increases business efficiency but simultaneously undermines the foundation of mass consumption.