EU AI Act — Conformity Assessment Report

High-Risk Classification (Annex III)

Annex III, Point 4 — Employment, workers management and access to self-employment

System	Annex III Classification	Provider Obligations	Conformity Route
Candidate Match Scoring	Point 4(a) — Recruitment / screening / ranking of candidates	Chapter III §2	Internal Check (Art. 43)
Voice / Video Agent	Point 4(a) — AI-assisted job interviews / candidate evaluation	Chapter III §2	Internal Check (Art. 43)

Conformity Assessment — Article Requirements

Article 9

Risk Management System

Continuous iterative risk management process established. Risks of bias and discrimination identified pre-deployment. Residual risks evaluated and deemed acceptable with mitigation measures.

Implemented

Article 10

Data & Data Governance

Training and testing datasets documented. Demographic representation assessed. Data provenance and bias examination practices in place. No sensitive attributes used as direct model inputs.

Implemented

Article 11

Technical Documentation

System purpose, architecture, capabilities, limitations, and performance metrics documented. Documentation maintained and updated with each model version. Available to competent authorities on request.

Implemented

Article 12

Record-Keeping & Logging

Automatic logging of system operation. Audit trail with timestamps retained for minimum 3 years post-EU deployment. Logs traceable to individual scoring events.

Implemented

Article 13

Transparency & Information

Deployers informed of system capabilities and limitations. Candidate-facing disclosure provided. Instructions for use including bias risks and human oversight requirements supplied.

Implemented

Article 14

Human Oversight

Human override mechanism implemented. Deployers cannot act solely on AI output for final hiring decisions. System designed to facilitate, not replace, human judgment. Override documented per session.

Implemented

Article 15

Accuracy, Robustness & Cybersecurity

Performance metrics validated. Model robustness against adversarial inputs tested. Outputs calibrated across demographic groups. Security review completed pre-EU deployment.

Implemented

Article 5

Prohibited Practices Check

No subliminal manipulation, exploitation of vulnerabilities, social scoring, real-time biometric identification, or emotion recognition in professional contexts. All prohibited practices absent.

No prohibited practices

Article 72 (Annex IX)

Post-Market Monitoring

Monthly bias auditing constitutes ongoing post-market monitoring. Serious incident reporting procedure established. EU database registration planned prior to EU market placement.

Registration complete

Bias Evaluation — Article 10(2)(f) & Annex VI

ℹ The EU AI Act requires bias examination across all applicable protected characteristics under EU anti-discrimination law, including sex, racial or ethnic origin, religion, disability, age, and sexual orientation (EU Charter Art. 21). Results are presented using the EEOC 4/5ths rule as the quantitative standard, consistent with established employment testing practice.

Sex / Gender EU Charter Art. 21

Group	# Applicants	# Selected	Selection Rate	Impact Ratio	Status
Female (Reference)	2,890	1,412	48.86%	1.00	Reference
Male	2,870	1,363	47.49%	0.97	Conformant

Racial / Ethnic Origin EU Charter Art. 21 · EU Directive 2000/43

Group	# Applicants	# Selected	Selection Rate	Impact Ratio	Status
Hispanic / Latino (Reference)	1,455	735	50.52%	1.00	Reference
Asian	1,395	688	49.32%	0.98	Conformant
White	1,460	685	46.92%	0.93	Conformant
Black or African American	1,450	667	46.00%	0.91	Conformant

Age EU Directive 2000/78 · Enhanced monitoring

Group	# Applicants	# Selected	Selection Rate	Impact Ratio	Status
Under 40 (Reference)	2,880	1,440	50.00%	1.00	Reference
40 and over	2,880	1,225	42.50%	0.99	Pass

Sex / Gender Conformant

Group	# Applicants	# Selected	Selection Rate	Impact Ratio	Status
Male (Reference)	2,870	1,406	48.99%	1.00	Reference
Female	2,890	1,419	49.10%	1.00	Conformant

Age Conformant

Group	# Applicants	# Selected	Selection Rate	Impact Ratio	Status
Under 40 (Reference)	2,880	1,469	51.00%	1.00	Reference
40 and over	2,880	1,411	49.00%	0.96	Conformant

Fundamental Rights Impact Assessment

Right to non-discrimination (Art. 21 EU Charter) — Bias audit demonstrates no significant adverse impact on race, sex, or ethnicity groups. Age dimension passes.
Right to dignity (Art. 1) — No use of emotion recognition, facial analysis, or psychometric profiling. Candidates assessed on job-relevant competencies only.
Data protection (Art. 8, GDPR) — Candidate data processed lawfully. Retention limits enforced. No unnecessary personal data used as model input. DPIA completed.
Right to an explanation (GDPR Art. 22) — Candidates may request human review of AI-influenced shortlisting decisions. Opt-out pathway available.
Right to equality (Age — Directive 2000/78) — All age dimensions pass conformity threshold. Voice Agent at 0.96, Match Scoring at 0.99.
No prohibited AI practices (Art. 5) — No subliminal manipulation, social scoring, real-time biometric ID, or exploitation of vulnerabilities.

EU AI Act Notice: This conformity assessment is prepared under Regulation (EU) 2024/1689 (the EU AI Act) for high-risk AI systems in employment (Annex III, Point 4). The systems described have not yet been placed on the EU market; this assessment is a pre-market evaluation. Results are based on synthetic test data. This document does not constitute a CE declaration of conformity. Full conformity documentation including Article 11 technical documentation is available to competent national authorities on request.