Before Applying to an AI Job, Run This 5-Point Check

You open an AI job posting, and the pay range makes you pause. Job-related scams increased by 38 percent in 2023. The UK's Report Fraud service received more than twice as many job scam reports in 2024 as in 2022. Before you spend the evening tailoring your résumé, test whether the role is real. The fastest way to protect your time is to inspect the posting like a document, not a promise. Your evening is worth more than a vague title.

Real AI work has a shape

Before you judge the posting, picture the work. A real AI role usually sits inside a business process: a model that scores leads, a pipeline that cleans data, a support tool that drafts replies, or a reporting system that turns raw logs into decisions. Genuine openings are anchored to particular hiring managers, departments, budgets, and start dates. A hiring manager, team lead, or product owner should be identifiable. A budget should exist, even if it is not printed in the ad. A start date should be plausible. When those details are missing, the role may be a placeholder or a scam. A real posting has a shape. A fake one is built to mislead.

AI-generated or scam postings are often too broad to be tied to a real hiring structure. They can be reused in any industry and still sound plausible. A vague AI posting should get your skepticism, not your application.

The bait is broad, not clever

Fake postings often pair an attractive title and high pay range with a thin, generic account of daily work. The workflow is the part you can verify; the pay range is not. The more impressive the title, the more closely you should read the daily work before you apply.

Fraudsters often promote broad remote openings, including data-entry work, AI support, and remote coordination, on social platforms. When the role appears only in a feed, a DM, or a group chat, treat the channel as a problem. A public job board is a safer place to start than a private channel.

Five red flags mark a posting to skip

Use this list on the posting, not on your hopes. Each check is quick. If the posting fails any of these, do not apply unless you can verify the role through a separate, credible channel.

• No named owner, budget, or start date. Can you identify a hiring manager, department, budget, or start date? When the posting hides all of them, the role may not exist yet.
• Vague AI duties. If the work is only work with AI, or if the daily responsibilities could describe any broad support role, the job is not defined.
• No workflow ownership. The role should own a specific process, dataset, model, or customer outcome. If the posting cannot name one, skip it before applying.
• Tool-name bait. The posting names tools only as keywords, with no task attached. Use large language models is not a job.
• No proof requirement. The application should ask for a sample, case study, code repository, portfolio, or structured interview. No proof means no skill test.

The text alone is not the test

AI detection tools are not especially dependable for staffing because scammers often revise the text by hand. A neat posting can still be fake. A plain posting can still be real. The structure is the better test.

Verification does not require a background check. It requires a few public signals. Search the company name plus the job title. Check the company's careers page, a recruiter profile, or a public job board. When the posting only exists in a private message, a social feed, or a one-off email, the employer should prove it. Ask for the hiring manager's name, the team, the reporting line, and the interview format. A legitimate employer should answer. If the person rushes you, avoids specifics, or moves the conversation to a payment or a device, stop.

This week, pick one saved AI posting, run the five checks, and either apply, skip it, or ask for proof before applying. Then make the decision in writing before you move on.