AI at work · Canadian practical guide

How do I know if a job uses AI screening?

An application portal does not reveal the whole hiring process. Here is how to look for reliable evidence and ask useful questions.

Editorial illustration of a job seeker inspecting an application beside an abstract screening interface and human recruiter
Original AI-generated editorial illustration; a concept, not a photograph of a real hiring process.

What AI screening actually means

AI screening may involve software that interprets application information to produce a recommendation, assessment or ranking. Hiring systems can also use simple rules: for example, checking an answer to a required-qualification question. Those processes are not automatically the same thing.

An applicant tracking system, or ATS, is software used to manage applications. It may store resumes, schedule interviews, run searches or offer AI features. Seeing an ATS brand or uploading a resume to a portal does not establish which features that employer has enabled. The useful question is what happens to your application in this particular recruitment process.

Where to look before you apply

Read the entire posting, including the footer. Look for an explicit statement about artificial intelligence, automated assessment, candidate matching or ranking. Save the posting and the date, because the wording may change. Read the notice shown when you create an account or submit your application.

Next, examine the employer’s recruitment privacy notice and any assessment-provider notice. They may explain the information collected, how it is used, who receives it and where to ask questions. A broad statement that a company “may use AI” does not tell you that your application will be automatically rejected; ask for the specific stage and use.

  • Strongest evidence: a clear employer disclosure or direct answer about the role.
  • Useful context: a specific assessment notice describing automated recommendations or scoring.
  • Inconclusive signs: a portal, an automated email, a chatbot or a request for a recorded interview.
  • Not reliable proof: an immediate rejection or a third-party claim that every employer uses the same screening method.

Ontario’s AI disclosure rule: the important limits

As of January 1, 2026, Ontario’s ESA rules require covered employers to disclose in publicly advertised job postings when they use AI to screen, assess or select applicants. The requirement does not apply to employers with fewer than 25 employees on the posting date; the official guide explains how the threshold is counted. Definitions and other exceptions also matter.

This is an Ontario requirement, not a blanket rule for every Canadian vacancy. Coverage depends on the employer, the posting and the applicable employment regime. The disclosure itself does not promise a particular explanation of an individual decision, human review or an appeal. A missing statement is not sufficient evidence that AI is absent—or, by itself, that an employer broke the law. Use the official guide and regulation for the details.

Sources: Ontario: requirements for publicly advertised job postings ↗ · Ontario Regulation 476/24: rules and exemptions for job postings ↗

A short message you can send the recruiter

Keep the question neutral and specific. You are trying to understand how to present your qualifications and whether you need an accommodation, not accuse the employer of unfairness. A recruiter may need to check with the recruitment or privacy team.

For example: “Before I complete the assessment, could you confirm whether automated tools or AI are used to screen, score or rank applicants for this role? Which stage uses them, what information is considered, and is there a contact for questions or an accommodation request?”

If the answer is vague, ask a narrower follow-up: “Does the tool only organize applications, or does it recommend which candidates advance?” Do not assume that a human making the final decision means the earlier shortlist had no automated influence.

Prepare a readable application without trying to trick a system

Follow the employer’s file and format instructions first. As a practical default, use recognizable headings, selectable text and clearly dated experience. Check any fields populated from your resume and correct parsing errors before submitting. A well-designed document should also remain easy for a person to read.

Use the job’s terminology where it accurately describes your experience. If a posting asks for a tool you have used, name it and explain what you did with it. Do not add hidden text, repeat keywords mechanically or claim credentials you lack. There is no universal keyword score that guarantees an interview across hiring systems.

Answer mandatory questions accurately. A rejection can have many causes, including eligibility, role changes, timing or stronger applicants. It usually cannot tell you which tool, if any, influenced the decision.

If there is a video interview or online assessment

A one-way recorded interview is a format, not proof that a model analyses your face or voice. Ask what is evaluated, whether recording is required and whether an alternative or accommodation is available. Check the provider’s notice before sharing more information.

For a technical or accessibility barrier, contact the employer before the deadline, explain the barrier and ask about the process for an alternative. Keep the correspondence. AI Speed’s internet test can help check your connection before an interview, but it cannot tell whether the employer uses AI or certify the interview platform’s reliability.

What to record if you have concerns

Keep the posting, relevant notices, assessment instructions and your correspondence. Note what happened and when. Avoid assuming that an unexplained rejection proves discrimination or that a technical failure proves an AI decision.

For a specific privacy, accessibility or employment-rights concern, identify the authority that covers the employer and issue. Federal, provincial and sector-specific rules differ. Request information from the employer’s named contact first when appropriate; this guide is a process checklist, not a ruling on your legal rights.

Frequently asked questions

Do all Canadian employers have to disclose AI screening?

No nationwide rule is established by Ontario’s requirement. Ontario’s rule has employer and posting coverage limits. Read the official guidance for your jurisdiction and ask the employer about the specific process.

Is an ATS the same as AI?

No. An ATS manages applications and may include AI features, ordinary searches and rule-based checks. The employer’s configuration determines what is actually used.

Does an instant rejection mean AI rejected me?

No. It could reflect a required answer, a closed role, a workflow rule or another reason. Timing alone cannot identify the decision process.

Can I demand a human review?

Ask whether review or an alternative process is available. Do not assume Ontario’s posting disclosure rule itself creates an automatic right to human review; applicable rights depend on the situation.

Sources and how we prepared this guide

We checked these primary sources on September 30, 2026. Research findings and jurisdiction-specific rules are identified in the text. Checklists, example messages and prompts are AI Speed’s practical suggestions; they have not been validated as predictors of job security or hiring success. This is general information, not individualized employment or legal advice.

  1. Ontario: requirements for publicly advertised job postings ↗

    Official ESA guide, including AI disclosure requirements effective January 1, 2026 and coverage exceptions; checked September 30, 2026.

  2. Ontario Regulation 476/24: rules and exemptions for job postings ↗

    Primary regulation defining artificial intelligence, covered postings and the fewer-than-25-employees exemption.

See our editorial policy or send a correction.