Is AI making hiring better, or just faster?

We gave hiring to AI. Here's what happened.

The pitch was simple: faster screening, less bias, better hires. And honestly, the efficiency case is hard to argue with. AI hiring tools now process tens of millions of applications a year, cutting hours of manual review down to seconds. For a founder trying to build a team without a full HR function, that sounds like exactly what you need.

But here's the thing nobody put in the brochure.

The bias didn't go away. It just got faster.

AI tools learn from historical hiring data. Which means if your past hiring decisions were biased — and statistically, most were — the model learns those patterns and repeats them at a scale no human manager ever could. What a biased hiring manager might have done to 50 candidates, an AI does to 50,000.

Research from the University of Washington found that AI resume screening tools favoured white-associated names in 85% of cases. Black male candidates were disadvantaged in up to 100% of direct comparisons with white male candidates. That's not a rounding error. That's the system working exactly as designed, on the wrong inputs.

The bias also hides in plain sight. Zip codes correlate with race. University names correlate with class. Employment gaps are more common among women who took time to raise children. A model that scores on any of these — even without meaning to — can produce discriminatory outcomes while looking completely neutral on paper.

The lawsuits are not theoretical anymore.

Mobley v. Workday achieved class certification in May 2025, covering potentially millions of job seekers over 40 who claim Workday's AI screening tool discriminated based on race, age, and disability. Courts are now asking whether AI vendors themselves can be held liable — not just the companies using the tools.

iTutorGroup paid $365,000 to settle the EEOC's first AI screening discrimination lawsuit after its software automatically rejected female applicants over 55 and male applicants over 60. HireVue faced criticism after its video interview AI disadvantaged non-native English speakers and neurodiverse candidates by scoring them on accents, facial expressions, and background noise. Most candidates didn't know any of this was happening.

If you're using an AI hiring tool right now, you're operating inside this liability environment whether you know it or not.

The regulations are a mess — and they're multiplying.

There's no single US law governing AI in hiring. Instead, you get a patchwork that's actively growing.

New York City already requires independent annual bias audits for automated hiring tools. California's new regulations require proactive bias testing, four years of records, and human override capability. Colorado's AI Act, effective June 2026, explicitly includes hiring tools as high-risk systems. Illinois and Texas have their own frameworks, both live as of January 2026.

A tool that's compliant in one state can create legal exposure in another. If you're recruiting nationwide, you're navigating multiple frameworks simultaneously.

Cross-border hiring makes all of this significantly harder.

Under GDPR, candidates in the EU already have the right to demand a human review of any AI rejection. They can ask what drove the decision. Most AI tools can't answer that question — because the model doesn't explain itself, and that silence is a direct GDPR violation.

EU regulators have issued 193 GDPR fines specifically in the employment sector, totalling over €360 million. US states introduced over 400 AI-related bills in 2024 alone. If you're hiring across borders, you're navigating frameworks that don't align, don't reference each other, and in some cases actively conflict. Your AI vendor built their tool for one market. The gap between markets is yours to close.

So what should you actually do?

You don't need to stop using AI in hiring. You need to use it with your eyes open.

Ask your vendor three things: What data was this tool trained on? What bias testing has been done, and can I see the results? What happens when a candidate requests an explanation for an automated rejection? If the answers are vague, the risk is yours — not theirs.

Make sure a human with actual authority to override AI decisions is part of your process. In a growing number of jurisdictions, that's not optional. And if you're hiring across borders, get compliance advice specific to each market before you deploy any automated screening.

The bottom line

AI made hiring faster. It also made it riskier in ways that weren't obvious until the lawsuits started arriving. The companies that come out ahead aren't the ones that abandoned these tools — they're the ones that understood what those tools were actually doing.

The regulations are tightening. The cases are accelerating. If you hire across borders, the complexity multiplies with every market you enter.

Don't wait for a complaint to find out where your exposure is. Talk to us before it becomes a problem.