Founding Batch

January 2027 is priced at founding rates. Fees rise 30% from Batch 2

Apply at Founding Rates
Zrooth
Menu
← All Insights

Insight · The Shift

AI Layoffs: The Problem Isn't Your Job, It's an Obsolete Skill Set

10 November 2026 · 8 min read

AI layoffs are mostly about tasks, not people. When a large employer cuts roles while reorganising around AI, the work itself usually remains; what disappears is the full-time role built on tasks that software can now do well enough. The fix is a new skill set, not a new job title.

This article reads the pattern behind the headlines, explains which roles shrink first and why, and sets out what a working professional, an aspiring founder or an SME owner in India, the UAE, Kenya or anywhere online can do about it. It does not predict which company cuts next, and it does not offer a shortcut. It offers a clearer way to look at what is happening.

What Is Actually Happening in AI Layoffs?

Over the past two years, a steady stream of large employers, multinationals among them, have announced job cuts while explaining that they are reorganising around AI. The headlines read as though AI is taking jobs. Look at which roles go first, though, and a more useful picture appears.

The roles that shrink are the ones built mostly on tasks a machine can now do well enough: drafting routine documents, summarising reports, first-line support, basic analysis, standard code. The company still needs that work done. It simply no longer needs a full-time person whose whole day consists of those tasks.

The research points the same way. The International Labour Organization's 2025 refined index of occupational exposure to generative AI found that one in four workers worldwide is in an occupation with some exposure, but only 3.3% of global employment falls into the highest exposure category. Its conclusion is that the transformation of jobs, not their outright automation, is the most likely impact, because most occupations still contain tasks that need human input (ILO, 2025).

Why Clerical and Routine Roles Feel It First

The same ILO study found that clerical occupations continue to have the highest exposure. That fits the task view. Clerical work is mostly reading, sorting, copying and drafting information against known rules, which is exactly the kind of work current tools handle. A role made up almost entirely of such tasks has little left once they are automated. A role that mixes them with judgement, client knowledge and responsibility keeps most of its value.

Why "AI Is Taking Jobs" Is the Wrong Frame

Seen this way, AI layoffs are a symptom rather than the cause. What AI is taking is a set of tasks, and a skill set that was valuable because those tasks were slow and scarce. When the tasks become cheap, the skill set built around them goes out of date. The person is not obsolete; the toolkit is.

This matters because the two frames lead to different responses. If you believe AI took your job, the natural response is to look for a similar job somewhere AI has not arrived yet, which is a shrinking target. If you see that AI took some of your tasks, the response is to move your time towards the tasks that remain valuable and to learn to run the automated ones yourself.

Augmentation Before Automation

How people actually use AI supports the task view. Anthropic's Economic Index, published in February 2025 from anonymised usage of its Claude assistant, found that use leaned towards augmentation (57%) rather than automation (43%), and that roughly 36% of occupations saw AI used in at least a quarter of their associated tasks (Anthropic, 2025). In other words, AI is mostly reshaping parts of many jobs rather than replacing whole jobs, which is exactly why the content of a role, not its title, decides how exposed it is.

What Do Employers Want Instead?

The same companies that cut roles are often hiring at the same time, for a different profile: people who can decide what to automate, run a process end to end with AI tools, judge the output, and answer for the result. Business adoption is broad enough that this profile is needed almost everywhere. The Stanford HAI AI Index 2025 reports that 78% of organisations said they used AI in 2024, up from 55% the year before (Stanford HAI, 2025).

Adoption on that scale creates a practical gap. Buying a tool is easy; making it change how a business runs is not. The skills that close that gap are business skills more than technical ones:

  • Reading how a business runs: where work comes in, who handles it, where it waits and where money leaks.
  • Choosing what to automate: which step is worth changing, and which is not worth the risk.
  • Judging tools honestly: testing them on real tasks rather than trusting demos.
  • Building and checking: assembling a working tool without code and testing whether it does what it should.
  • Costing the change: putting a defensible number on the process before and after.
  • Taking responsibility: setting the rules an AI system works within and owning the outcome.

None of these is new. What is new is that they used to sit with senior managers and consultants, and now they are the everyday work of anyone who runs a process with AI in it.

How AI Layoffs Look in India, the UAE and Kenya

The pattern behind AI layoffs is global, but it lands differently by market, so it is worth being specific about where the answer depends on place.

India

India has a large services workforce in IT, back-office operations and support, which are exactly the areas built on routine, well-documented tasks. That makes the task shift highly visible. It also means India has a large pool of people who already understand how these processes run from the inside, which is the knowledge needed to redesign them with AI.

The UAE

In the UAE, many professionals work in regional headquarters, finance, real estate and services firms that adopt technology quickly. The shift there tends to show up as smaller teams running more with AI, and as demand for people who can set up and manage that way of working.

Kenya and the Wider African Market

In Kenya and across Africa, many businesses are still building their operations, which can be an advantage: a firm that has not yet hired a large back office can design its processes around AI from the start. The opportunity is less about defending an existing role and more about building lean operations early. We explore this further in building AI-native operations from Nairobi.

What Should You Do If Your Role Is Exposed?

The useful response to AI layoffs is practical and starts with your own week, not with the headlines.

  1. List your tasks, not your title. Write down what you actually did over a typical fortnight. Mark each task as routine (could be done by a tool with clear instructions) or judgement-based.
  2. Find what remains valuable. The judgement, the client knowledge, the understanding of how your industry works: that is your real asset. Our companion piece on which skills AI makes obsolete and which it rewards gives a simple test for this.
  3. Learn the tools on a real process. Pick a process you know well and try to run part of it with AI. One working example teaches more than a course certificate.
  4. Learn to cost it. Put a number on the process before and after, as set out in costing a process before and after AI.
  5. Keep a record of the work. What you built, what it changed and how you checked it is stronger evidence than a list of courses.

If you have already been affected, the steps are similar but the starting point is different; see laid off because of AI? how to rethink your skills.

Where These Skills Are Actually Learnt

These skills cannot be learnt from a syllabus alone. They are learnt by working on real businesses: mapping where a business leaks time and money, building a working tool without code, and knowing what you can and cannot delegate to an agent. Reading about them helps; doing them on a live process is what makes them stick.

A few honest cautions. Learning these skills takes real effort over weeks, not a weekend. They do not remove the risk of restructuring, and no course or programme can promise that a particular employer will value them. What they do is move your time towards the part of the work that is growing rather than shrinking.

How the Programme Approaches This

That is the premise of the AI Entrepreneurship Programme: twelve weeks of business consulting and training, with eight weeks of guided training and four weeks of remote practical work with an assigned assistant. It runs in Navi Mumbai, Dubai, Abu Dhabi and Nairobi, and online, on in-office, weekend hybrid and online tracks. You can see the sequence in what you learn and how it works, or try Pathfinder to see how it would apply to your field.

The programme is training. It does not promise a job, a placement or any outcome; it replaces an out-of-date skill set with one that lets you run, or build, a business with AI.

Sources

Frequently Asked Questions

Are AI layoffs really caused by AI?

Partly. Companies often cite AI when restructuring, and other reasons such as cost pressure usually play a role too. The useful question for you is which of your tasks a tool can now do, because that is what makes a role easier to cut.

Which jobs are most exposed to generative AI?

The ILO found in 2025 that clerical occupations have the highest exposure. More broadly, roles made up mostly of routine reading, drafting, sorting and summarising are more exposed than roles that mix those tasks with judgement and responsibility.

Will AI replace my whole job?

For most roles, research points to transformation rather than replacement: some tasks are automated and the rest change. The risk is highest when nearly all of a role consists of routine tasks a tool can now do.

Is it too late to retrain if I am mid-career?

No. Mid-career professionals often hold the most valuable part of the new skill set: knowledge of how a business and an industry really work. What they usually need to add is hands-on practice with the tools.

Do I need to learn to code to stay relevant?

No. AI now writes competent code for common tasks. What is scarce is the ability to decide what is worth building, specify it clearly and check whether it works, which does not require programming.

How are AI layoffs different in India, the UAE and Kenya?

The task shift is the same, but its effect varies. India has large services and back-office sectors, the UAE has fast-adopting regional firms, and many Kenyan businesses can design lean, AI-assisted operations from the start.

What evidence should I show employers about my AI skills?

A short record of real work: the process you changed, how you tested the tool, what it cost before and after, and what you would do differently. That is more persuasive than a list of course certificates.

Does the AI Entrepreneurship Programme lead to a job?

No. It is business consulting and training, and it does not offer or promise a job, a placement or any income. You leave with a record of the work you did, which you can use however you choose.

Pathfinder · Free

Where Do You Fit in the AI Economy?

Type your industry and the work you do. See the live businesses you would learn inside, your twelve weeks, and three venture ideas made for you.

Find My Path

Join the Discussion

Agree, disagree or add your own experience. Every comment is reviewed before it appears; spam and abuse are removed, and disagreement is welcome.

    Add Your View

    0 / 2,000
    Links send a comment to manual review. Please do not share personal details about other people.