Laid off because of AI: what to do in the first thirty days
A first-month sequence for people whose role was cut in an AI-related restructuring — the psychological part first, then runway, then the search — with the mistakes that make month two harder.
If your role was cut and the reason given — or the reason everyone assumes — was AI, you are dealing with two things at once: the ordinary loss of a job, and a specific story about why, one that says the loss is about your obsolescence rather than a company's spreadsheet. The second part is what makes this harder than a normal layoff, and it is also the part that is mostly untrue.
Week one: process the event, don't research it
There is a counter-intuitive finding in the job-insecurity research: a definite loss is often easier on mental health than prolonged uncertainty, because it can be processed. You have a definite event. Let it be one. That means: tell the people who need to know, take the administrative steps (severance terms, benefits continuation, unemployment claim on day one where applicable), and give yourself a defined few days of not doing anything strategic.
What not to do in week one: read every article about AI and your former field. That is not research; it is rumination with a browser. It will not tell you anything actionable and it will attach the layoff to a story about your obsolescence that you will then have to un-learn.
Week two: the runway number
Accessible savings plus severance, divided by your monthly floor — the amount needed to keep housing, obligations and health, not your current spending. Write it down. This number determines the pace of everything else, and knowing it converts a formless dread into a specific problem with a deadline. If it is under two months, the priority is emergency support and any bridging income before anything else; the guide's runway chapter lists the routes. If it is longer, you have time to run a proper search rather than a panicked one.
Week two, also: correct the story
Do the three-boxes exercise (title, craft, qualities). A restructuring cut a title. It did not cut your craft — the things you concretely know how to do — and it did not touch your qualities. When you write the CV and the application, you are selling boxes two and three; the title that was cut is the least of what you have. Most people who do this exercise after an AI-related layoff find that the "obsolete" story fits about a quarter of what they actually do.
Weeks three and four: the search, framed properly
Search at the task level, not the title level. Which roles need the judgement, relationships and context you have, regardless of what they are called? Adjacent roles in other industries are often the fastest route back, because the craft transfers and the market for it is different. Use the tools in the search itself — visibly, in the application — because "I know what these tools do and I know what they don't" is now a credential rather than a threat.
Pace: a small, fixed number of well-targeted applications per day beats a large number of scattershot ones, both for outcomes and for your state of mind. And schedule the day. Unstructured time is where rumination lives.
The mistakes that make month two harder
Taking the first offer out of fear rather than fit — it costs you a year. Not claiming the support you are entitled to because it feels like admitting something. Isolating, because telling people is embarrassing; the people you tell are the search. And letting the sleep go, because "I'll sleep when I've found something" is precisely backwards: the search is worse on broken sleep. The 3am rule and the worry appointment apply here as much as anywhere.
When to get help
A layoff is a recognised risk period for depression, and the AI framing adds an identity wound to it. If the low mood is most of the day most days for more than two weeks, if you have stopped doing things, or if there is any thought of not wanting to be here, that is the point to talk to someone — US 988, UK 116 123, findahelpline.com elsewhere — and the low-cost routes exist precisely for people between jobs.
Frequently asked
What should I do first after being laid off because of AI?
Administrative steps on day one — severance terms, benefits, unemployment claim — then a defined few days of processing the event without researching it. In week two, calculate your runway and do the three-boxes exercise to separate what was cut (a title) from what you still have (craft and qualities).
Does being laid off because of AI mean my skills are obsolete?
Almost never in full. Restructurings cut roles; they do not assess individuals task by task. When people list what they concretely do and mark each task as production or judgement, the obsolescence story usually fits a fraction of the list. The rest transfers, often to adjacent roles in other industries.
How do I explain an AI layoff in interviews?
Briefly and without the story: the company restructured and the role was eliminated. Then move to what you can do, with the tools mentioned as something you use rather than something that happened to you. Interviewers care far more about the second half.
Is it normal to feel depressed after an AI layoff?
Low mood after job loss is common and a layoff is a recognised risk period for depression. If it is most of the day most days for more than two weeks, or comes with loss of interest or hopelessness, talk to a professional — low-cost routes exist specifically for people between jobs.
Sources
- Dekker, S. W. A., & Schaufeli, W. B. (1995). The effects of job insecurity on psychological health and withdrawal: A longitudinal study. Australian Psychologist, 30(1), 57–63.
- Sverke, M., Hellgren, J., & Näswall, K. (2002). No security: A meta-analysis and review of job insecurity and its consequences. Journal of Occupational Health Psychology, 7(3), 242–264.
- De Witte, H. (2005). Job insecurity: Review of the international literature on definitions, prevalence, antecedents and consequences. SA Journal of Industrial Psychology, 31(4), 1–6.
- O*NET OnLine (U.S. Department of Labor) — occupation task data used in automation-exposure estimates; free.