When Your Staffing Site Screens a Candidate It Shouldn’t

A candidate on your recruiting website mentions a disability that needs accommodation and asks whether their age might be a problem for the role. The chatbot, built to qualify leads efficiently, factors it in and tells them they are probably not a fit. In one exchange the site asked about protected characteristics, made a screening decision that touched them, and put the whole thing in writing. That transcript is not a lead record. It is the outline of a discrimination claim, and the staffing firm owns it.

Staffing might be the riskiest place of all to drop a general-purpose chatbot. The core activity is deciding who advances, and that single decision sits under the densest body of anti-discrimination law in all of employment. A bot built to screen does the one thing that creates the most exposure when it is done carelessly. It just does not know that.

Screening Is Where Discrimination Law Lives

The moment a chatbot evaluates candidates and decides who moves forward, it is making employment decisions. Those decisions run headlong into Title VII’s prohibition on discrimination based on race, color, religion, sex, and national origin. The danger is rarely a bot announcing bias out loud. It is disparate impact. A screening pattern quietly filters out a protected group, or the bot questions people inconsistently and treats two similar candidates differently. An automated screener applying criteria no one audited can produce that at volume, in a record a plaintiff’s attorney or the EEOC can read line by line.

Some Questions Are Off-Limits, and the Bot Will Ask Them

A chatbot built to gather context will wander into territory employers are trained to avoid. Anything touching disability runs into the ADA’s limits on disability-related inquiries. Age, family status, religion, national origin: a recruiter learns to step around every one of them. A bot has no such instinct. It asks whatever helps it "qualify" the candidate. So a friendly question about whether someone can "handle the physical demands," or "has childcare sorted out," becomes evidence the firm collected and weighed something it should never have touched.

AI Hiring Tools Are Now Specifically Regulated

This is no longer general principle. Regulators have turned directly to automated hiring. New York City now requires bias auditing and disclosure for automated employment decision tools. Other jurisdictions are following. The EEOC has made clear that reaching for AI does not excuse a firm from anti-discrimination law. A staffing company running a screening chatbot may already sit inside rules written for exactly this technology, and "the vendor’s tool did it" is not a defense any of them accept.

Background Checks Add the FCRA

When screening touches background or credit information, the Fair Credit Reporting Act’s rules govern how those reports may be used and what a candidate is owed before and after an adverse decision. A chatbot that rejects someone based on that information without the required process skips protections the candidate is legally entitled to, on top of whatever discrimination exposure the screening already created.

"Will Not" Is a Suggestion. "Cannot" Is an Architecture.

A bot built to qualify candidates will ask and weigh the very things that create liability. Gathering information is its default. A prompt is only a request to hold back. That is the gap between a tool told to avoid protected topics and one built so it cannot screen on them at all. The second kind routes an accommodation request to a human and follows a compliant adverse-action process. A disclaimer beneath the chat does not help when the transcript shows the firm asked and decided.

Who Owns the Answer

A recruiter who asked about a candidate’s disability and age and then screened them out would be a severe training failure, and staffing firms drill anti-discrimination discipline because the exposure is enormous. An unsupervised chatbot does it at scale, in writing, with no one reviewing the questions or the decisions, and the liability does not transfer to the software vendor. It stays with the firm, now holding a record that reads like a discrimination case and may already violate rules written specifically for AI hiring tools.

The firms that get burned are not the ones that modernized intake. They are the ones that let a bot screen candidates and ask the questions recruiters are trained never to ask. The fix is an assistant that captures interest and answers questions about open roles and the process. Anything past that, screening, an accommodation request, any decision that touches a candidate, goes to a trained human working under a compliant process.

Frequently asked questions