Should you use AI to write your job applications?

It is widely accepted to use AI to write job applications when that means help with your own material. Handing over the whole job is not. That is roughly what the evidence says, and the rest of this page is the detail.

You would not be alone either way. A recruiting firm's 2025 market report, as summarised by the US Chamber of Commerce's small-business publication, found about two-thirds of candidates say they use AI when applying, and nearly 40% think it both helps and hurts their search.^[1] That mixed feeling is reasonable.

The question worth asking is narrower than yes or no. Which parts of an application can a tool help with, and which parts have to come from you? Facts about your work, your reasons for wanting a job, and the decision to send it are yours. Wording, order and spelling are where help is welcome.

Some will. The same article reports a resume-service survey of 600 US hiring managers:^[1]

• Nearly 20% said they would reject a candidate who used an AI-generated resume or cover letter.
• 14.5% thought candidates should not use AI at any stage.
• 52% said it is acceptable to use AI for proofreading or support in drafting documents.

One survey, from a company that sells resume writing, is not the last word. But the shape is consistent with what employers publish themselves: help is fine, substitution is not.

A growing number publish them. The Scottish Government's candidate guide is a clear example. It says your application should represent your skills and personal experiences accurately and authentically. It allows AI to help refine and clarify your ideas, to research public information, and to check spelling, grammar and clarity. It rules out using AI to inflate or invent skills and experiences, to create generic responses and copy them into the application, or to answer questions in an interview or live assessment. Where plagiarism is identified, the application is withdrawn.^[4]

That is one employer, in one country. Policies differ, and some are stricter. So the first practical step is simple: before you apply, look on the careers page or in the posting for a line about AI. If there is one, follow it.

There is real research on this, and it cuts both ways.

One study looked at an AI cover-letter tool introduced on a large online labor platform. Access to the tool made cover letters match job posts more closely and raised callback rates. Time spent editing the AI draft was positively correlated with being hired. But after the tool arrived, the link between a well-matched cover letter and a callback fell by 51%, and employers put more weight on other signals, such as workers' track records.^[2]

A second study, by economists at Dartmouth's Tuck School and Princeton, examined a freelance marketplace before and after AI writing tools. Before, a carefully tailored proposal signalled effort and ability. Afterwards, that signal stopped predicting who did the job well. The researchers estimate that workers in the top fifth of ability were hired 19% less often, and those in the bottom fifth 14% more often.^[3]

Both studies are about freelance work, not salaried jobs, so read them as a direction and not a forecast. The direction is this: when everyone's writing is polished, polish stops counting. What still counts is what is true about you, and what you add by hand.

Sometimes, and less reliably than they think. In the survey above, a third of hiring managers (33.5%) said they can quickly detect a resume created with AI.^[1] That is their own estimate. Nobody tested it.

The safer way to think about it: a reader does not need to prove anything. An application that sounds like the last forty gets passed over, whatever wrote it. An application with a specific project, a number and a plain sentence gets read, whatever helped write it.

It is worth knowing the screen runs in both directions. The same article says more than 60% of companies use AI somewhere in recruiting, about 20% of them for resume screening.^[1] The public is uneasy about that. In a Pew Research Center survey of 11,004 US adults, 66% said they would not want to apply to an employer that used AI to help make hiring decisions.^[5]

A working method, drawn from the above:

1. Write the facts yourself first. What you did, for whom, with what result. Rough notes are fine.
2. Ask for structure, not content. Let the tool order your notes, cut repetition and fix grammar. These are the uses employers' own policies allow.^[4]
3. Edit the draft. In the cover-letter study, time spent editing went with better outcomes.^[2] Replace anything you would not say out loud.
4. Check every claim. If the draft says you led something you only joined, fix it. You will be asked about it.
5. Do not paste the same generic answer everywhere. It is the one use that both the policy and the surveys treat as a problem.^[1][4]
6. Send it yourself. Read the whole form once before you press Submit.

If the employer's policy says no AI at all, skip steps two and three for that application and write it yourself.

None of this guarantees a reply. It keeps the application honest and recognisably yours, which is the part you control.