AI & hiring

AI in Recruitment: Transparency Statement

Version 1.0 · August 2026 · For employers using RaydarIQ and their counsel

1 · Purpose and scope

This statement describes how RaydarIQ's AI candidate-screening agent works, what it does and deliberately does not do, and how candidate data is handled. It is intended to support an employer's AI-tool inventory, job-posting disclosure obligations (including Ontario's Employment Standards Act posting requirements, in force January 1, 2026), and internal legal review.

It covers the screening agent — the AI that conducts structured phone screens with job applicants. RaydarIQ's other capabilities (roleplay certification, knowledge checks, call analysis) are employee training and coaching tools: they involve people already hired, not applicants, and sit outside hiring-decision regulation. They follow the same design principles described here.

2 · What the screening agent does

An applicant calls a screening line at a time of their choosing (or, where the employer has enabled it and consent allows, receives a call after applying). An AI voice agent conducts a structured interview: logistics questions defined by the employer (eligibility to work, availability, transportation, start date) and situational questions about experience, communication, and goals. Every applicant for a role receives the same questions. The call is transcribed and scored against a rubric the employer defines, producing a scorecard: dimension scores, strengths, flags with reasons, and coaching notes — with the full transcript and recording attached. The scorecard is delivered to the employer's recruiting team.

3 · What the screening agent does not do

4 · Human oversight

Every scorecard is reviewed by the employer's human recruiter, who decides who advances — and can overrule any flag ("fast-food-only work history" is a flag; the recruiter who values eager first-jobbers advances the candidate). The disposition — who is hired, who is not, and why — is made and communicated entirely by the employer's people. Any candidate may request a human-conducted screen instead of the AI at any point, including mid-call.

5 · Candidate notice and consent

Candidates are informed of AI screening at the point of application (notice language is agreed with each employer). The call itself opens with disclosure — the live script begins:

"Hi, thanks for calling [Employer]. My name is Alex, and I'm an AI recruiting assistant... Just so you know, this call is being recorded for our recruiting team to review. If you're not comfortable with that, just let me know and we can end the call, and someone from our team will reach out to schedule a live call instead."

For Ontario posting compliance, an accurate disclosure line for a posting using RaydarIQ screening is: "We use artificial intelligence to conduct and summarize initial phone screenings. All hiring decisions are made by people."

6 · Data handling

Collected: the applicant's identity and contact details as provided, their interview responses (audio, transcript), and the derived scores and notes. Processed by: established service providers in four categories — telephony, speech transcription, AI analysis, and secure database hosting (specific sub-processor list available on request; processing may occur in the United States under contractual safeguards). Retained: per the retention schedule agreed with each employer, after which recordings and transcripts are purged. Access: limited to the employer's authorized recruiting staff and RaydarIQ operational personnel. Audit trail: every screen retains its complete record — questions asked, responses, scores, and the recruiter's decision — supporting the employer's record-keeping obligations (three years under Ontario's ESA) and any future review.

7 · Fairness and consistency

The strongest fairness property of structured AI screening is uniformity: every candidate gets the same questions, evaluated against the same rubric, with the same patience, at any hour — no interviewer fatigue, no drift between candidate one and candidate forty. Employers define the scoring criteria, and RaydarIQ works with each employer to keep those criteria job-related. Scoring behavior is monitored on an ongoing basis, and the per-candidate audit trail (section 6) means any screen can be re-examined end to end.

8 · About this document

This statement is informational and is not legal advice; it is intended for review by each employer's counsel, and RaydarIQ will answer any further diligence questions directly. Contact: tom@raydariq.com.