An AI Tool Drafted the Job Ad. The Employer Still Owned the Legal Risk

The Justice Department reached a settlement with Elegant Enterprise-Wide Solutions after the company posted job advertisements that restricted consideration to applicants with H-1B, OPT, or H-4 status. The advertisements were generated by an AI tool. DOJ’s position was unambiguous: responsibility for the content remained with the employer.

The matter ended in a $9,460 civil penalty and three years of compliance obligations, including policy review, training for personnel involved in recruiting or approving job ads, and record preservation. It was not a court judgment, but the settlement reflects DOJ’s determination that using generative AI does not reduce an employer’s legal exposure.

Drafting automation is still part of the hiring process

Many organizations treat job-ad drafting as low-risk compared with screening or interviews. The Elegant settlement shows why that view is too comfortable. A job posting defines who is invited to apply. A restriction in the ad can exclude workers before any application is reviewed.

Generative tools can introduce unlawful preferences in several ways: through the prompt, through inference from prior examples, or through vague instructions like “make this more targeted.” None of those paths changes who published the advertisement.

Human review must be specific

A policy requiring “human review” of AI-generated content is only effective if the reviewer knows what to look for. A useful review should test the posting against the actual legal or contractual basis for any restriction, check whether the language is broader than necessary, and compare the generated version against the approved template.

Reviewers should also see what changed. If a system silently rewrites previously approved language, a final read can miss new limitations.

Employers need provenance for job-ad text

When a challenged advertisement appears, the company should be able to reconstruct how it was created. That record should include the original template, the prompt or inputs given to the AI, the generated draft, the approver, and the platforms and dates on which the ad appeared.

Without that provenance, the company may know what was published but not why the language appeared or who could have caught it.

The practical rule

Automated drafting does not create automated immunity. If a company publishes a job advertisement, it should be ready to explain the source of any restriction, show who approved it, and demonstrate that its review process was designed to catch unlawful language.

An AI tool can draft the text. The employer still owns the decision to use it.