Skima AI Releases Responsible Hiring AI Transparency Report

Report documents Skima AI’s bias evaluation methodology, human oversight model, data governance controls, and planned third-party audit path for AI-assisted candidate screening.

OAKLAND, California – SP Tech Solutions – July 31, 2026 — Skima AI today released its Responsible Hiring AI Transparency Report, documenting how its AI-assisted candidate screening system is evaluated for potential adverse impact, how hiring decisions remain under human control, and how candidate data is handled within its screening architecture.

The report is intended for HR, recruiting, talent acquisition, legal, compliance, and procurement teams reviewing AI tools used in hiring workflows. It outlines Skima AI’s fairness evaluation process, including disparate impact analysis using the Four-Fifths Rule, synthetic profile testing, intersectional analysis, and protected class review across sex, race and ethnicity, age, disability status, and compound demographic groups.

According to the report, Skima AI evaluated its core candidate-screening model across 792 profiles and 5 demographic splits. All evaluated demographic groups returned impact ratios above the 0.80 Four-Fifths Rule threshold, and no group breached the concern threshold in the internal evaluation. Skima AI states that full methodology documentation and raw evaluation data are available to enterprise clients under NDA.

The report also describes the product controls Skima AI uses to limit the role of automation in hiring decisions. Skima AI’s system ranks, scores, and surfaces candidates against job-related criteria, but does not make final hiring decisions. Candidate progression, rejection, and offer decisions require authorization from a human recruiter or hiring manager.

The transparency report includes:

• Internal bias evaluation results across sex, race and ethnicity, age, disability status, and intersectional combinations
• Impact ratio calculations based on the Four-Fifths Rule methodology
• Scoring signal controls focused on job-related skills, experience, qualifications, and tenure patterns
• Protected attribute exclusion from candidate scoring inputs
• Human-in-the-loop controls for screening, shortlisting, interview review, and final hiring decisions
• Data processing architecture for candidate resumes and personally identifiable information
• Client documentation availability for enterprise legal, audit, and procurement review
• Planned an independent third-party bias audit aligned with automated employment decision tool review expectations

“AI-assisted hiring systems need documented evaluation, not broad claims,” said Dinesh Chaudhary, Head of Machine Learning at Skima AI. “This report gives enterprise teams a clearer view of how Skima AI evaluates screening outputs, limits scoring signals to job-related criteria, and keeps final hiring decisions with human reviewers.”

The release of the report comes as employers and software providers face increased scrutiny over automated tools used in hiring and workforce decisions. In the European Union, AI tools used for employment, worker management, and access to self-employment are treated as high-risk use cases under the EU AI Act framework. In the United States, employers using algorithmic decision-making tools must consider whether selection procedures create an adverse impact under applicable employment discrimination laws.

Skima AI’s report states that the deploying employer remains responsible for final hiring decisions and applicable legal compliance. Skima AI’s role is to provide screening architecture, documentation, evaluation data, and audit-support materials that help enterprise teams review and govern AI-assisted hiring workflows.

The Responsible Hiring AI Transparency Report is available on Skima AI’s Bias Evaluation page. Enterprise clients may request additional methodology documentation, raw impact ratio data, architecture diagrams, and audit-support materials under NDA.

About Skima AI

Skima AI is an AI recruitment platform that helps hiring teams search, screen, match, and rediscover candidates across recruitment workflows. The platform is designed to support recruiters and hiring teams by ranking candidates against job-related criteria, surfacing relevant candidate evidence, and integrating with existing recruitment systems while keeping hiring decisions under human review.

Media Contact

Priyanshu Dhiman
SEO Lead
Skima AI
press@skima.ai
https://skima.ai