AI & Responsible AI Policy

1. AI features covered

• Candidate Intelligence and the Hiring Score
• CV analysis and candidate/job matching
• AI-assisted evaluation of screening answers and Career Snapshots
• AI summaries and recommendations
• ATS Resume Match and AI JD Generation, where available

2. Human decision-making

AI features support recruiters and do not replace human decision-making. Recruiters remain responsible for final hiring decisions and should review the underlying evidence, not just the score.

3. About the Hiring Score

• It is job-specific: a score for one job says nothing about another.
• It uses only the available job and candidate information.
• It may contain uncertainty and may be affected by incomplete information.
• It is not a guarantee of future job performance.
• It must not be the sole basis for any hiring decision.

Where available, each score is broken down into dimensions with a short rationale, so recruiters can see the evidence behind it.

4. What we do not evaluate

Our AI features are designed not to evaluate candidates on inappropriate or protected characteristics, including:

• Race or ethnicity
• Religion
• Gender
• Age
• Disability
• Physical attractiveness or facial appearance
• Facial expression or emotion inferred from appearance
• Voice or accent
• Any other sensitive or protected characteristic

Career Snapshot information must only be used for legitimate, job-relevant purposes, such as what a candidate says about their experience.

5. Limitations

• Incomplete information in a CV, Snapshot or job description
• Incorrect matching of skills or experience
• Hallucinations - confident but wrong statements
• Outdated information
• Missing context that a human would notice
• General limitations of the underlying models

If an output looks wrong, rely on your own review. You can re-run a score or disregard it.

6. Data and model training

AI model providers process job and candidate information only to generate outputs for your workspace. The current AI provider is OpenAI - gpt-6-astra. BlinkCV does not use Customer Data or Candidate Data to train general-purpose AI models. AI processing is performed to provide the applicable BlinkCV functionality and in accordance with the applicable privacy, contractual and data-processing terms.

7. Recruiter responsibilities

• Write job descriptions and screening questions that are job-relevant and non-discriminatory.
• Review AI-generated content before publishing it.
• Give candidates a fair, human review.
• Report concerns about AI behaviour to info@blinkcv.ai.

8. Contact

Questions about this policy can be sent to info@blinkcv.ai.