AI Terms & Transparency Notice

This notice explains how RareFourth's AI-powered interview and evaluation works, so candidates and Clients understand what the system does, what it does not do, and what rights and safeguards apply. It supplements our Privacy Policy and Terms & Conditions.

1. How the AI works

• Interviewing. An AI interviewer selects and asks questions based on the role and adapts follow-ups in real time to your responses. It does not provide scores, feedback, or hints to the candidate during the interview.
• Evaluation. Your answers, code, and (where applicable) transcribed speech are assessed by AI models against a structured rubric to produce dimension-level scores, a summary, potential red flags, and a hire/no-hire recommendation.
• Proctoring. Camera images and integrity signals are analyzed by an AI vision system to detect indications of additional people, devices, or screens, and by rule-based checks for behaviors such as tab switching or pasting. Suspicious observations may be captured as evidence.

2. What the AI does not do

• It does not make the hiring decision. It produces a recommendation for the Client, who decides.
• It does not claim certainty. Detections (including proctoring flags) are probabilistic indicators, not conclusive proof of any conduct.
• It is not infallible and may produce errors or omissions. Outputs should be reviewed by a human before any decision that affects you.

3. Human oversight

Every evaluation is delivered to a human at the Client, who can review the questions, your answers, the scoring rationale, and any evidence before acting. We design the system to support — not replace — human judgment in hiring.

4. Fairness and limitations

We use a consistent rubric applied to every candidate to promote fairness and reduce individual bias. However, AI systems can reflect limitations in their design or data. We work to monitor and improve accuracy and fairness, and Clients remain responsible for using outputs lawfully.

5. Your rights regarding automated processing

The AI produces a recommendation; the hiring decision is made by a human at the Client. Under the DPDP Act, 2023, you may access a summary of the personal data we process about you, seek correction or erasure, and raise a grievance. To request human review of an evaluation or raise a concern, contact our Grievance Officer at [privacy@rarefourth.com], and we will route it to the responsible Client where appropriate.

6. Data used by the AI

The AI processes only the interview and proctoring data described in the Privacy Policy. Candidate interview content is not used to train third-party foundation models. We may use aggregated or de-identified data to evaluate and improve our own system.

7. Consent

Camera-based proctoring involves sensitive personal data under the SPDI Rules, 2011 and is carried out only with your prior explicit consent, captured before the interview begins. You may decline, though the interview may not proceed, and you may withdraw consent as described in the Privacy Policy.

8. Contact

Questions about how our AI works or to exercise your rights: [privacy@rarefourth.com] (Grievance Officer: [Name]).