AI recruiting is legal but increasingly regulated. In the US, New York City's Local Law 144 requires bias audits of automated hiring tools, and the EEOC warns that AI hiring must not create discriminatory outcomes. The EU AI Act classifies recruitment AI as high-risk. Compliant AI hiring keeps a human in the loop, uses explainable scoring, and documents how candidates are evaluated.
Yes, using AI in recruiting is legal in most jurisdictions, but it is not unregulated, and the rules are tightening. There is no blanket ban on AI hiring tools; instead, existing anti-discrimination law applies to them just as it does to human decisions, and a growing set of AI-specific regulations layers additional obligations on top. The practical reality is that you may use AI to source, screen, and assess candidates, provided the tools do not produce discriminatory outcomes and you meet the transparency, audit, and human-oversight requirements that apply where you operate. The legal risk is not in using AI per se but in using it carelessly — deploying a tool you cannot explain, that has not been checked for bias, or that makes consequential decisions with no human review. Compliance, not prohibition, is the operating question.
New York City's Local Law 144, in effect since July 2023, is one of the most specific AI-hiring rules in the United States. It regulates 'automated employment decision tools' used to screen candidates for jobs or promotions in the city. Employers using such a tool must have it subjected to an independent bias audit within the prior year, publish a summary of the audit results, and notify candidates that an automated tool is being used and what characteristics it assesses, giving them a chance to request an alternative process. The law is significant because it moves beyond general anti-discrimination principles to impose concrete, auditable requirements. If you hire in New York City with AI-driven screening, meeting these bias-audit, disclosure, and notice obligations is not optional — it is a specific legal duty with enforcement behind it.
The US Equal Employment Opportunity Commission has made clear that federal anti-discrimination law applies fully to AI-driven hiring. Through technical guidance, it has warned that automated selection tools can violate Title VII if they produce a disparate impact — disproportionately screening out candidates by race, sex, or other protected characteristics — even where there was no intent to discriminate. It has also addressed the Americans with Disabilities Act, cautioning that AI assessments can unlawfully disadvantage candidates with disabilities if they are not accessible or if they screen for traits unrelated to the job. The EEOC's position is essentially that employers remain responsible for the outcomes of any tool they use, including a vendor's, so 'the software did it' is not a defence. Testing tools for adverse impact and ensuring accessibility are the expected safeguards.
The European Union's AI Act, which entered into force in 2024, explicitly classifies AI systems used in employment — including tools that screen or filter job applications and evaluate candidates — as 'high-risk.' That classification carries substantial obligations for providers and deployers: risk management, high-quality training data to limit bias, technical documentation and record-keeping, transparency to those affected, human oversight, and appropriate accuracy and robustness. In practice, deploying recruitment AI for candidates in the EU means you cannot treat it as an ordinary software purchase; it sits in a regulated category with compliance duties that phase in over time. For any company hiring in Europe, understanding whether a tool and its use fall under the high-risk regime, and ensuring the vendor supports the required documentation and oversight, is now part of due diligence.
Beyond the headline regimes, several other laws bear on AI hiring. Illinois' Artificial Intelligence Video Interview Act requires employers using AI to analyse video interviews to notify candidates, explain how the AI works, and obtain consent. Colorado has enacted a broad AI law addressing algorithmic discrimination in high-risk systems, including employment, with obligations set to take effect in 2026, and other US states are advancing similar bills. In Europe, the GDPR's Article 22 gives individuals rights around decisions based solely on automated processing that have significant effects, which is directly relevant to fully-automated hiring rejections. The landscape is fragmented and moving quickly, so the safest posture is to track the specific jurisdictions where your candidates are located rather than assume a single national standard covers you.
Compliant AI recruiting shares a consistent set of practices across these regimes. Keep a human in the loop for consequential decisions rather than letting the tool auto-reject candidates unreviewed. Use explainable scoring so you can articulate why a candidate was ranked or filtered a certain way, which supports both fairness and disclosure duties. Test tools for adverse impact — the bias audits Local Law 144 mandates and the EEOC expects — and document those checks. Be transparent with candidates that AI is used and, where required, obtain consent and offer alternatives. Ensure accessibility for candidates with disabilities. Maintain records of how the system evaluates people. Together these amount to a principle: use AI as decision support with oversight, evidence, and disclosure, not as an unaccountable black box that makes hiring decisions on its own.
The employer generally bears the liability, even when the tool comes from a third-party vendor. Anti-discrimination law focuses on the outcome and the party that made the employment decision, so an employer cannot typically shift responsibility to the software provider by pointing to the vendor's algorithm. This is why the EEOC stresses that using a vendor's tool does not relieve you of the duty to ensure it does not produce discriminatory results. Vendors may share exposure and increasingly face their own obligations under laws like the EU AI Act, but for the hiring employer the practical takeaway is ownership: you are accountable for what the tool does in your process. That makes vendor due diligence — asking for bias-audit results, documentation, and evidence of fairness testing — a legal safeguard, not just a procurement nicety.
Get a personalized walkthrough of Pitch N Hire on your own roles and workflow. No slides, no obligation.
Prefer to talk? Book a demo · View pricing
Free 1-user plan · No credit card · Talk to a real hiring expert
See how Pitch N Hire automates sourcing, screening and AI interviews on your real roles. Start with your work email — no credit card.
★ Free 1-user plan · No spam · Talk to a real hiring expert
