How to Tell If AI Screened Your Resume — and What to Do Next
Did a machine reject your resume before a person ever saw it?
Often, yes — but "AI screened it" can mean three very different things, and only one of them decides whether you advance.
Figures and rules in this post are current as of November 2026. Hiring law is changing fast, so treat the legal details as a starting point, not legal advice.
First, know which "AI screening" you are dealing with
Most of what job seekers call AI screening is sorting, not deciding.
Your application passes through software before a recruiter opens it. What that software does falls into three levels, and they matter very differently for your chances:
- Parsing: The system reads your resume into fields — name, job titles, dates, skills — so a recruiter can search it. A badly parsed resume is harder to find, but nothing has rejected you.
- Ranking or scoring: The system scores your resume against the job description and orders the pile. A low score can bury you below the fold, but a person still makes the call.
- Automatic rejection: The system itself removes you. This usually happens for one of two reasons: a knockout question you answered the "wrong" way, or a score below a set cutoff.
The distinction matters because your response changes at each level. Parsing problems are fixed with simpler formatting. Ranking problems are fixed with the posting's own language. Automatic rejection is usually fixed by answering the knockout questions correctly — or by asking the employer what happened.
We mapped the three steps in detail in How AI Screens Your Resume Before a Human Sees It. What follows is how to spot which one you are facing, at each stage of the process.
Before you apply: the posting often tells you
The cheapest check costs thirty seconds and happens before you upload anything.
- Ontario postings: Since January 1, 2026, Ontario employers with 25 or more employees must state in a publicly advertised job posting whether AI is used to screen, assess, or select applicants. If you are applying for an Ontario role at a larger employer and the posting says nothing, that silence is itself a signal — either no AI is used, or the employer is not following the rule.
- Disclosure language: Search the posting for "automated", "AI-assisted", "algorithmic assessment", "automated screening", or "bias audit". Employers that use these tools increasingly name them, because in several jurisdictions they are now required to.
- New York City roles: Under Local Law 144, an employer using an automated employment decision tool must publish a summary of its most recent bias audit on its careers site and notify candidates at least 10 business days before the tool is used. Search the careers page for "bias audit" — if you find one, AI screening is confirmed before you apply.
- Vendor names: Watch for names like HireVue, Pymetrics, Eightfold, or Paradox in the posting, the application portal, or the privacy notice. A vendor name is the most concrete evidence you will get, because it tells you exactly which system is involved.
Save a copy of the posting before you apply. If the process later changes — a resume review becomes a recorded video interview, for example — you will have a clear record of what you were told at the start.
While you apply: the portal gives itself away
Watch the address bar and the questions, not the branding.
- Where the application lives: If clicking "Apply" sends you to a domain such as myworkdayjobs.com or boards.greenhouse.io, you have entered an applicant tracking system. The application is being parsed and stored before any human sees it.
- "Powered by" labels: Scroll to the bottom of application forms for "Powered by" labels and check the browser tab for platform names such as Workday, Greenhouse, SuccessFactors, or iCIMS. These platforms increasingly include AI features that rank candidates and suggest shortlists.
- Knockout questions: Yes-or-no questions about work authorization, a required license, or minimum years of experience are the one mechanism that genuinely auto-rejects. They are rules a person configured, not a model judging your prose — and one wrong answer can end the application on the spot.
- One-way video or voice interviews: If you record answers with no live interviewer present, and the instructions mention analysis, scoring, or assessment of your responses, AI is evaluating you. In Illinois, the AI Video Interview Act has required notice, an explanation, and consent before AI analyzes a video interview since 2020.
- Consent screens and match scores: Read the consent screen instead of clicking through it. Language about automated decisions, profiling, or analysis of your voice and video is a direct disclosure. Some portals even show you a match score — that number is the ranking step, made visible.
After you submit: timing tells the story
A rejection that arrives in minutes was not read by a person — but it probably was not read by a model either.
Instant rejections, within minutes or a few hours of applying, almost always trace to a knockout answer or a score below a cutoff. The most common explanation is a knockout question: a missing license, a "no" on work authorization, or a salary expectation outside the configured range. The second is a keyword or score threshold your resume did not clear.
A status change to "no longer under consideration" within hours points the same way. A human did not make that call at 2 a.m.
Weeks of silence usually mean something else entirely: your resume is sitting in an unviewed queue. That is not a rejection, and it is not proof of AI involvement either. It is the normal state of a pile a recruiter has not reached yet.
And a generic rejection email proves nothing on its own. Automated and human rejections use the same templates. Timing and the application steps you went through are far better evidence than the wording of the email.
Where the law already forces employers to tell you
Four jurisdictions now require some form of disclosure — and each one works differently.
New York's rule is the most useful for a job seeker because the bias-audit summary is public: you can read it on the employer's careers site before applying. Illinois's rule sits inside the state's Human Rights Act, so a failure to give notice is a civil rights violation in itself — and its detailed notice rules were proposed in May 2026 and withdrawn in June 2026, while the statute's notice duty still applies. Colorado's comprehensive AI law, by contrast, has its enforcement paused while amendments are debated.
None of these laws guarantees a human will read your resume. What they give you is a paper trail — and a question the employer is now required to answer.
What to do once you suspect AI screened you
Knowing is only useful if it changes what you do next.
1. Ask in writing
Email the recruiter or the hiring contact before you complete an assessment, or after a fast rejection. Keep it short and specific:
"Before I complete the assessment, could you confirm whether AI will analyze or score my resume, written answers, voice, or video — what decision that informs, and whether a person reviews the result?"
In New York City, you can also ask how to request an alternative selection process. The employer must tell you how to ask, though the rules do not force them to provide one. In Ontario, the posting itself should already have answered the question — if it did not, that is worth noting in your message.
2. Write for the machine that reads first
If a parser or a ranking step stands between you and a human, write for it first and for the human second. Use plain formatting with no tables, columns, or graphics that can corrupt parsing. Mirror the posting's exact terms for skills and qualifications, and put must-have requirements where the parser can find them — in your experience bullets, not only in a summary.
Our checklist for writing a resume each machine reads correctly is in How AI Screens Your Resume Before a Human Sees It. And if you use AI tools to draft the application itself, the line between help and harm is covered in AI Tools for Job Seekers: What to Hand to AI and What to Check Yourself.
3. Keep a record
Save the job posting, the consent screen, and the rejection email with its timestamp. Screenshot the portal pages that name a vendor or show a match score. If a required notice never arrived in New York City, Illinois, or Ontario, that record is what a complaint to the regulator — or a conversation with an employment lawyer — would rest on.
Key Takeaway
AI screening is rarely one invisible judge. It is a chain: a parser reads your resume, a ranking step orders the pile, and only sometimes does a rule or a score remove you without a human. Each link in that chain leaves evidence — in the posting, in the portal, and in the clock.
If you remember one thing: check the posting before you apply, watch the portal while you apply, and read the clock after you submit — then ask in writing, because in more places than job seekers realize, the employer is now required to answer.
Sources:
- • NYC Local Law 144 Explained: AI Hiring Bias Audits (2026) — Archuz
- • Illinois Adopts New AI-in-Employment Regulations — Hinshaw
- • Ontario AI Hiring Disclosure: Employer Guide — SortTheStack
- • EU AI Act Hiring Rules and Applicant Rights — Upleva
- • How to Tell If AI Is Screening or Scoring Your Job Application — UMA Technology
- • AI in Hiring: What You Are Entitled to Know — ResumeVera

Comments
Post a Comment