AI and Work: The Good, the Bad, and the Unemployed

AI’s impact on work is uneven: some jobs are assisted, some are strained, and some are left behind.

By: Rick

"The future of work is not one story but several happening at once."

Artificial intelligence is often sold as a workplace productivity tool. Sometimes it is. Sometimes it is a cost-cutting tool dressed up as progress. Quite often, it is both at once. Reuters has reported this month (April 2026) that companies including Snap, Meta, and others are linking restructuring and job cuts to AI-driven efficiency and investment shifts, while Deloitte says the most successful organizations are not merely adding AI tools but redesigning roles, workflows, and career paths around them.

That is why the impact of AI on work does not come as one story. It comes as at least three.

Some work becomes easier.

Some work becomes worse.

Some work disappears.

The good

There is no point pretending AI has no upside at all. It clearly does.

In many workplaces, AI can reduce drudge work, speed up routine tasks, assist with drafting, summarize large amounts of information, support software development, and help employees move faster through repetitive steps. Deloitte reports that improving productivity and efficiency is the top benefit organizations say they are getting from enterprise AI adoption, with two-thirds reporting gains. Deloitte also says some organizations are seeing 30–35% productivity gains across parts of the software development life cycle.

That is significant. Anyone who has spent years buried in emails, routine documentation, repetitive research, or procedural tasks can understand the appeal immediately. There is genuine value in tools that lift some of that weight.

There is also a more hopeful version of this story. AI can help people do more ambitious work by taking some of the mechanical load off their shoulders. Deloitte argues that the strongest long-term use of AI is not just faster output, but a rethinking of work that combines human strengths with AI capabilities.

So yes, there is real good here. Some jobs may become less tedious. Some workers may become more capable. Some organizations may genuinely free people for better work.

The bad

But this is where the glossy version usually stops.

Productivity gains are not the same thing as better work. Sometimes “efficiency” just means fewer people doing more, under tighter pressure, with AI inserted into the middle of already strained systems. Reuters’ recent coverage of AI-related restructuring shows that the business case is often not only about helping workers. It is also about leaner teams and lower labor costs.

That creates a familiar pattern. AI removes some drudgery, but it can also add new burdens: checking weak outputs, correcting mistakes, supervising machine-generated material, adapting to unstable workflows, and absorbing the expectations that everything should now happen faster because “AI is helping.”

There is also a subtler problem. Work does not only give people income. It gives rhythm, competence, identity, apprenticeship, and a path upward. When AI strips away too much of the intermediate work, especially the beginner work, it can hollow out the ladder people once climbed. Deloitte notes that many organizations are still educating employees about AI, but far fewer are fully re-architecting roles, workflows, and career paths. That gap may become a significant problem as these situations evolve.

So the “bad” is not simply that AI makes mistakes. It is that it can degrade the texture of work itself: less interesting and challenging jobs, less training, less patience for learning, more surveillance of output, and more pressure to keep pace with systems that never tire. The last clause is an inference, but it follows from the productivity and restructuring logic described across current reporting.

The unemployed

This is the part people feel in their gut, and not without reason.

The World Economic Forum’s Future of Jobs Report 2025 says employers expect 170 million jobs to be created globally by 2030 and 92 million displaced, for a net gain overall, but that broad number does not make the losses less real for the people and sectors hit first. The same WEF framing stresses that the central challenge is transition: how intentionally and sustainably organizations manage workforce transformation.

Reuters has also reported that economists at Goldman Sachs estimated AI was responsible for 5,000 to 10,000 monthly net job losses last year in the most exposed U.S. industries, and Reuters’ recent layoff coverage shows companies actively citing AI efficiency and AI-focused restructuring as part of current job cuts.

That is why the reassuring line, “AI will create new jobs,” is only partly comforting. Historically, major technological shifts often do create new work. But the people who lose existing jobs are not automatically the same people who can step into the new ones, and the timing is rarely neat. WEF’s own discussion emphasizes that the issue is not just the total number of jobs, but the ability to navigate the shift into new roles.

This is especially worrying for white-collar and entry-level work. If AI can draft, summarize, analyze, sort, code, and coordinate at acceptable quality for many routine tasks, then some of the very jobs people once used to enter professional life may shrink first. That point is an inference from current AI capability trends and the restructuring pattern described by Reuters and Deloitte, not a direct claim from one source.

The deeper truth

The future of work is probably not a clean one-to-one swap in which a human disappears, and an AI takes over the desk.

The deeper change is reorganization.

Deloitte says the most successful organizations are reimagining jobs to combine human strengths and AI capabilities, and it points to emerging roles such as AI operations managers, human-AI interaction specialists, and quality stewards. Its 2026 human capital material likewise argues that success will depend on how organizations redesign workflows, governance, and culture as AI becomes a collaborator in everyday decisions.

That means the workplace is not just facing replacement. It is facing redistribution. Some roles will be thinned. Some tasks will vanish. Some workers will become supervisors of AI systems. Some organizations will cut headcount and call it innovation. Some will genuinely create better hybrids between human judgment and machine speed.

The result will not be one future but several, unfolding unevenly across sectors, classes, firms, and countries. That unevenness is one reason the current conversation often feels so confused: people are arguing about different parts of the same wave.

The real question

The real question is not whether AI will affect work. It already is.

The real question is what kind of workplace we are building around it.

Are we using AI to remove drudgery and expand human capability?

Are we using it to intensify pressure and thin out jobs?

Are we preparing people for transition, or simply congratulating ourselves that new jobs may exist somewhere, for someone, later on?

AI can make work better. It can also make work lonelier, thinner, and more precarious. And for some people, it will remove work altogether.

That is why the conversation about AI and work cannot be reduced to optimism or panic. It has to remain moral, economic, and human at the same time.

Because the future of work is not only about what AI can do.

It is about what employers choose to do with it.