Save your time
for the right candidates.
Pre-cruit is candidate screening software which lets you filter unqualified candidates and get video answers for your interview questions.
Priya N.
Senior Support Lead
Your team's scores
reviewed by a human ✓
The case against AI-scored hiring
Peer-reviewed research, audits & litigation summaries. Full citations below.
- 85.1% — of the time, White-associated names were favoured over equally qualified others.
- 100% — of Black male candidates lost in every case tested.
- 94% — résumé summaries leaked protected attributes like pregnancy or political status, unprompted.
- 1.4% points — Black male candidates faced up to 1.4-percentage-point lower chance of being "hired" by ChatGPT than equally qualified White men — a gap that, scaled across just the US labor force, translates to nearly 150,000 jobs negatively affected.
- 21% — GPT-4 ranked a résumé with disability-related honors (recognition/awards mentioning disability) first in only 15 out of 70 trials against its disability-free twin
- 0.85 — correlation between AI's gender bias and existing job gender gaps.
- 5/5 — AI models tested all showed racial or gender bias.
- 86.67%–100% — of the time, ChatGPT selected the very first résumé presented out of ten equally qualified candidates — versus the 10% you'd expect from pure chance.
- 60% — Candidates were up to 60% more likely to be shortlisted when their résumé was written by the same AI model doing the screening — even with identical content quality.
The fix everyone's selling is worse than the problem.
Too many applicants
Easy-apply and AI-polished résumés mean the pile is bigger and the signal is weaker than ever. Manual review doesn't scale, and your week disappears into screening the unqualified.
AI as the “solution”
Nearly every alternative tool leans on AI to filter, assess, rank or score candidates. It doesn't fix hiring — it adds bias, legal risk, and candidates who quit halfway.
We're not guessing. Here's the receipts.
Three reasons AI scoring fails hiring teams — each one backed by peer-reviewed research, active litigation, and real candidate behavior. Expand any card to see the sources.
0%
of the time, AI résumé screeners favored white-associated names.
Audit after audit finds the same thing: large language models pick up the prejudices in their training data. They are biased by gender, race, disability, pregnancy status, even the order applications arrive in or whether the resume was written by the same AI that is screening.
- Wilson & Caliskan, “Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval” — AIES 2024
- “Are Emily and Greg Still More Employable than Lakisha and Jamal?” — algorithmic hiring bias in the era of ChatGPT — 2023
- Puutio & Lin, “First Come, First Hired” — SSRN 2025
- Xu, Li & Jiang, “AI Self-preferencing in Algorithmic Hiring” — arXiv 2025
- FAIRE benchmark — 2025
- Armstrong et al., “The Silicon Ceiling: Auditing GPT’s Race and Gender Biases in Hiring”
- “Measuring gender and racial biases in LLMs: intersectional evidence from automated résumé evaluation” — PNAS Nexus
- Glazko et al., “Identifying and Improving Disability Bias in GPT-Based Resume Screening” — ACM FAccT 2024
- “Emotion AI in Job Interviews” — ACM FAccT 2025
- “Person Perception Biases Exposed: Revisiting the First Impressions Dataset” — HBU 2021
- HBS — “Hidden Workers: Untapped Talent”
0 landmark cases
and a wave of new laws now target automated hiring decisions.
HireVue drew an FTC complaint and an ACLU discrimination complaint. Mobley v. Workday won collective certification under the ADEA. NYC Local Law 144 and Illinois' AI Video Interview Act add bias-audit and disclosure duties. Opaque AI scoring is exactly what regulators are coming for.
- EPIC — In re HireVue (FTC complaint & litigation)
- ACLU of Colorado complaint against HireVue — March 2025 — alleged discrimination against deaf and non-white candidates
- Mobley v. Workday — ADEA disparate-impact claim; collective certification, May 2025
- AI interview screening lawsuits — tracker
- NYC Local Law 144 — bias-audit + disclosure for automated employment decision tools
- Illinois Artificial Intelligence Video Interview Act
- Manshel — “The Bigotry of the Future: AI Recruitment Tools & Hiring Discrimination Law” (CUNY Law)
0 in 10
candidates drop out once they learn an interview is AI-led.
Only 26% of applicants trust AI to evaluate them fairly. The rest hesitate, disengage, or walk — and recruiters admit AI tools are screening out strong people. The candidates you most want to reach are the ones most likely to leave.
Pre-cruit: the signal, minus the AI judge.
See and hear how people actually think — then let your team decide. No scoring models, no transcribing-to-rank, no black box.
no applicants yet?
Build a pool from scratch
Share a public interview link on your job post, careers page, or socials. Candidates qualify and record themselves — you end up with a shortlist-ready pool of video answers.
already have applicants?
Invite them in two clicks
Add the people you already have and send interview invites by email. They get a link, record their answers, and drop straight into your review queue.
Build your interview once
Add qualifying questions to filter fast, then video questions for the answers that matter. Set thinking time, retakes, and ideal length. Done in minutes.
Candidates record on their own time
One link, any schedule, just Google Chrome. No app, no account, no calendar tetris. They answer when they’re at their best.
Your team reviews and decides
Watch answers side-by-side with the question. Score holistically or per criterion, leave comments, shortlist or pass. Every call is human.
You set the bar.
You move the slider.
Two ways to score — both fully human. Watch each answer next to its question, leave comments your team can see, and shortlist with a click. There is no model in the loop, anywhere.
One holistic score
Give each candidate a single 1–5 rating after watching their answers. Perfect for high volume or a quick gut-check across the team.
- 1–5 per candidate
- Great for volume
- Zero setup
Criteria per question
Define what matters — communication, ownership, role skills — and score every answer against the same bar. Consistent across reviewers, defensible later.
- Your rubric, your weights
- Same bar for everyone
- Per-question detail
Faster hiring, without the baggage.
Save your week
Replace phone screens and scheduling with answers you can review in minutes. One recruiter can clear a whole pipeline in a sitting.
Eliminate algorithmic bias
There is no model ranking, scoring, or flagging anyone. Humans watch real answers and decide — so the bias documented in AI tools never enters your pipeline.
Stay on the right side of the law
Laws like NYC LL144, Illinois’ AIVIA, and the EU AI Act target automated decision tools that score candidates. Pre-cruit makes no automated assessment — so it sidesteps that category.
AI scoring tools vs. Pre-cruit
* Pre-cruit performs no automated assessment of candidates. This is not legal advice - consult your counsel about your obligations.
The “but really, no AI?” section
No — and that is the whole point. Pre-cruit never scores, ranks, transcribes-to-judge, or flags candidates with AI. It captures video answers and gives your team the tools to review them. Every decision is made by a person.
Regulations like NYC Local Law 144, Illinois’ AI Video Interview Act, and the EU AI Act are aimed at automated employment decision tools that score or rank applicants. Because Pre-cruit makes no automated assessment, it falls outside that category. (This isn’t legal advice - check with your counsel.)
keep the judgment human
Skip the pile.
Keep your judgment.
Set up your first interview, share a link, and start reviewing real answers today. Your team decides — every single time.
No credit card · works in Chrome · candidates need no account