OpenAI pays $3.2M to settle DOJ hiring discrimination claims

OpenAI and its subsidiary Statsig will pay $3.2 million to settle Justice Department claims they discriminated against U.S. workers in hiring for PERM positions, per the Civil Rights Division's August 4 announcement. That splits into $1.2 million in civil penalties and $2 million in compensation for affected workers.

The alleged mechanics matter more than the dollar figure. Investigators say OpenAI left PERM roles off its public careers site, a labor-certification process tied to sponsoring foreign workers for permanent residency. For those specific jobs, the company required paper applications while everything else on its site took electronic submissions. Some job ads reportedly ran as radio spots late at night. Together those choices make a role technically postable while quietly hard for a U.S. applicant to find or apply to.

OpenAI denied wrongdoing but agreed to revise its policies, retrain staff, and submit to DOJ monitoring going forward. That last piece is the part worth watching. A fine gets paid and forgotten. A monitoring period means someone outside the company is checking whether the new process actually produces different outcomes, not just different paperwork.

This is the largest of 13 settlements to come out of the DOJ's relaunched Protecting U.S. Workers Initiative, which targets exactly this pattern: PERM postings structured so a foreign candidate already selected for sponsorship is the only realistic hire. For a company burning through visa-sponsored engineering talent as fast as OpenAI is, that pattern isn't unusual. Getting caught at this scale is.

What the DOJ's release doesn't say is whether any of the affected roles were ultimately filled by the intended sponsored candidates, or how the $2 million gets divided among the workers it's meant to compensate.