TIDEZINE.

DOJ Says OpenAI Buried Job Postings in Late-Night Radio Ads to Shut Out Local Applicants

The U.S. Department of Justice accuses OpenAI of deliberately hiding job openings, requiring paper-only applications, and running late-night radio ads to freeze out American job seekers in favor of visa-holding workers — a scheme that ended in a $3.2 million settlement.

DOJ Says OpenAI Buried Job Postings in Late-Night Radio Ads to Shut Out Local Applicants

If you're job hunting, late-night radio ads probably aren't where you'd expect to find your next opportunity. But that's exactly at the heart of the U.S. Department of Justice's (DOJ) allegations against OpenAI: one of the most closely watched AI companies in the world reportedly advertised certain job openings on late-night radio instead of posting them on its own public careers page.

According to findings released by the DOJ, OpenAI and one of its subsidiaries failed to broadly advertise certain positions through the standard Permanent Labor Certification (PERM) process — a process designed to ensure employers give U.S. workers fair consideration before sponsoring foreign employees for visas. The DOJ says that instead of listing these roles on external job boards, OpenAI required applicants to submit their resumes by mail, even though the company accepted electronic applications for all its other openings. Investigators concluded this pattern of behavior effectively locked American job seekers out of the running, clearing the path for workers on temporary work visas to fill those roles instead.

The case was resolved with a $3.2 million settlement. Of that, $1.2 million will go to the U.S. government as a civil penalty, while the remaining $2 million will fund back-pay compensation for job seekers who may have been unfairly passed over. The DOJ says it will review every application submitted by U.S. workers to determine whether they were genuinely given fair consideration at the time; the fund is intended for applicants who applied but weren't properly evaluated — not for people who simply never saw the ads and never had a chance to apply in the first place.

Under the settlement, OpenAI is now required to post open positions publicly on its own careers website, accept electronic applications, and train staff on anti-discrimination provisions under the Immigration and Nationality Act.

This isn't the first time a tech giant has been caught running this kind of playbook. In 2021, Meta reached a multimillion-dollar settlement with the DOJ following a lawsuit over its H-1B visa hiring practices. In 2023, the DOJ found Apple had used similarly low-profile recruiting tactics. Zooming out, U.S. scrutiny of H-1B visas has intensified sharply in recent years: in September 2025, President Trump signed an executive order imposing an additional $100,000 fee on certain H-1B visa applications, a move that's hit tech companies reliant on foreign talent to varying degrees. OpenAI's settlement is just the latest example of a tech company getting caught using old-school tricks to dodge the rules amid this tightening regulatory climate.

Related

$42B Net Loss and Another AI Extinction Warning—Anthropic Still Pushing for $2T IPO
Tech

$42B Net Loss and Another AI Extinction Warning—Anthropic Still Pushing for $2T IPO

Anthropic's IPO draft rarely admits in its own prospectus that its AI could pose an "existential risk" to humanity, while also revealing a $42 billion loss last year—yet its valuation could still reach $2 trillion.

Florida AG Files Emergency Injunction Against OpenAI, Demanding Halt to New Model Training and Block on Minor Users
Tech

Florida AG Files Emergency Injunction Against OpenAI, Demanding Halt to New Model Training and Block on Minor Users

Florida Attorney General Uthmeier filed an emergency motion on Monday asking the court to bar OpenAI from training new models without independent safety oversight and to cut off minors' access to ChatGPT, in a case stemming from a criminal investigation earlier this year into the FSU shooting.

OpenAI AI Agent Hacked Australian Government Website on Its Own, PM Blasts Company for Three-Month Reporting Delay
Tech

OpenAI AI Agent Hacked Australian Government Website on Its Own, PM Blasts Company for Three-Month Reporting Delay

Australian Prime Minister Anthony Albanese confirmed that an OpenAI AI agent breached the government's Medicare public health insurance system website back in June, but OpenAI didn't disclose the incident until September—with several similar undisclosed intrusions occurring in between.

Amazon's New Fire TV Stick 4K Touts Faster Boot Times, Slimmer Body, While New Remote Goes by Feel
Tech

Amazon's New Fire TV Stick 4K Touts Faster Boot Times, Slimmer Body, While New Remote Goes by Feel

Amazon has launched its next-gen Fire TV Stick 4K, claiming boot and launch speeds 20% to 40% faster than similarly priced competitors. It's priced at $60, dropping to $30 during Prime Big Deal Days.

Xbox Launches Mythic Achievements, Finally Scratching That Decade-Long Platinum Trophy Itch
Tech

Xbox Launches Mythic Achievements, Finally Scratching That Decade-Long Platinum Trophy Itch

Xbox has unveiled a new system called Mythic Achievements, awarded to players who fully complete a game's original achievement list. The system retroactively applies all the way back to Xbox 360-era titles, though it's currently only available to Xbox Insider testers.

TikTok Pays $100M to Settle Lawsuit—All It Buys Is a 2-Hour Daily Cap for Minors
Tech

TikTok Pays $100M to Settle Lawsuit—All It Buys Is a 2-Hour Daily Cap for Minors

TikTok has settled a social media addiction lawsuit with the state of Alabama, agreeing to pay at least $100 million and rolling out measures targeting underage users—including a two-hour daily usage cap and a ban on beauty filters. Failure to follow through could mean an additional $300 million fine.