The Predictive Index
HireAIScore rates The Predictive Index 48 out of 100 (grade F, Substantial gaps) under rubric v1.0, last reviewed July 17, 2026. The Predictive Index ranks 29 of 82 vendors rated in Assessment AI, against a category average of 46. That score is drawn from 24 evidence items across 7 rubric criteria for The Predictive Index — 18 with a cited source, 6 recording that nothing was located.
Score breakdown
The Predictive Index, criterion by criterion
Each rubric criterion is scored out of 100 on the cited evidence, then weighted into the total.
Weighted contribution shown to the right of each bar.
Article 11 Technical Documentation
Weight 20% · category avg 50
50
+10.0
Bias Audit Transparency
Weight 18% · category avg 41
45
+8.1
FRIA Support
Weight 15% · category avg 31
35
+5.3
Data Governance Disclosure
Weight 15% · category avg 53
55
+8.3
Human Oversight Design
Weight 12% · category avg 55
52
+6.2
Post-Market Monitoring
Weight 12% · category avg 38
43
+5.2
Customer Documentation
Weight 8% · category avg 58
60
+4.8
Total weighted score
47.8(rounded to 48 for display)
Category avg is the mean raw score on that criterion across the 82 Assessment AI vendors in scope of this rubric, this one included.
Strongest criterion
Customer Documentation
60raw / 100
Contributes 4.8 to the total. 60 against a 58 category average.
Weakest criterion
FRIA Support
35raw / 100
Contributes 5.3 to the total. 35 against a 31 category average.
Cited evidence
The evidence behind the score
Every score is backed by at least one cited piece of evidence. Open a criterion to see what was found, when, and where.
Download diligence recordEvidence ledger
- Items
- 24
- Documentation
- 14
- Audit report
- 1
- Public statement
- 3
- Absence
- 6
- With a source URL
- 18 of 24
- Source hostnames
- 2
- Fewest items
- 3
- Bias Audit Transparency
These figures measure how thoroughly The Predictive Index was reviewed, not how The Predictive Index performed — an absence row, recording that nothing was located, is counted like any other item.
Article 11 Technical Documentation
4 items50
- DocumentationCaptured Jul 17, 2026
Public research-and-validity page documents the assessments' validation approach, differential item functioning analysis, EFPA certification, and ongoing psychometric monitoring, but states detailed technical documentation is available only via a PI consultant.
https://www.predictiveindex.com/learn/support/research-and-validity-at-pi/
- DocumentationCaptured Jul 17, 2026
The AI ('Obi') page provides an explainability/principles statement describing quality-control steps (PI-validated core plus Rubric-Based Verification) and design principles, functioning as an AI policy statement rather than formal Article 11 technical documentation.
- DocumentationCaptured Jul 17, 2026
Assessment-validity page cites 400+ job-performance studies and documents fairness via DIF studies, but offers no downloadable technical pack or model card.
https://www.predictiveindex.com/assessments/assessments-validity/
- AbsenceCaptured Jul 17, 2026
No AI system/model cards, no ISO 42001 certification, and no publicly downloadable technical documentation pack were found on the PI site or Trust Center.
Bias Audit Transparency
3 items45
- Audit reportCaptured Jul 17, 2026
Both assessments underwent independent third-party EFPA (European Federation of Psychologists' Associations) certification review covering fairness and validity, described as PI's largest peer review and redone every three years, though the certificate itself is not published.
https://www.predictiveindex.com/learn/support/legality-and-compliance-of-pi-assessments/
- DocumentationCaptured Jul 17, 2026
PI reports its own differential item functioning analyses found only small or negligible gender, race, and age effects, but these are internal (non-independent) analyses without a published dataset.
https://www.predictiveindex.com/learn/support/introduction-to-fairness-bias-and-adverse-impact/
- AbsenceCaptured Jul 17, 2026
No published NYC Local Law 144 bias audit, and no independent AEDT adverse-impact audit (e.g., BABL AI, ORCAA, Warden AI, Holistic AI) were found via web search or on PI's site.
FRIA Support
3 items35
- DocumentationCaptured Jul 17, 2026
PI provides general deployer guidance to conduct job analyses and adverse-impact analyses across the selection process, but this is EEOC-oriented and not an EU AI Act FRIA or deployer-obligation framework.
https://www.predictiveindex.com/learn/support/understanding-adverse-impact/
- DocumentationCaptured Jul 17, 2026
A Vanta-hosted Trust Center (confirmed via network requests) provides a trust package with policies and documents available on request, but contains no FRIA template or AI Act deployer guidance.
- AbsenceCaptured Jul 17, 2026
No EU AI Act Article 27 Fundamental Rights Impact Assessment template or explicit Article 26 deployer-obligation guidance was found.
Data Governance Disclosure
4 items55
- DocumentationCaptured Jul 17, 2026
PI confirms certification to the EU-US, UK Extension, and Swiss-US Data Privacy Framework Principles via the U.S. Department of Commerce.
https://www.predictiveindex.com/privacy/data-privacy-framework/
- Public statementCaptured Jul 17, 2026
The AI page asserts SOC 2 and GDPR compliance and states user data is never used to train third-party AI models nor shared externally with third-party providers.
- DocumentationCaptured Jul 17, 2026
Vanta-powered Trust Center (verified via assets.vanta.com network requests) exposes subprocessors, resources, and security documentation, with document downloads gated behind an access request.
- DocumentationCaptured Jul 17, 2026
PI publishes a GDPR data processing addendum available to customers on request through its legal department.
Human Oversight Design
3 items52
- Public statementCaptured Jul 17, 2026
PI positions its AI as 'built to support, not replace,' framing outputs as speeding human decision-making rather than automating it.
- DocumentationCaptured Jul 17, 2026
PI instructs deployers that no single assessment alone should be used to make a hiring decision, reinforcing human-in-the-loop use.
- AbsenceCaptured Jul 17, 2026
No documentation of concrete override controls, decision audit logs, in-product AI explainability, or per-jurisdiction compliance toggles was found.
Post-Market Monitoring
3 items43
- DocumentationCaptured Jul 17, 2026
The Vanta Trust Center includes updates, subprocessors, and security-contact surfaces (confirmed via network requests to Vanta's trust-updates and trust-subprocessors modules).
- Public statementCaptured Jul 17, 2026
PI states its assessments are constantly monitored by its Science Team to ensure strong psychometrics and fairness, indicating ongoing (though non-public-dashboard) monitoring.
https://www.predictiveindex.com/learn/support/research-and-validity-at-pi/
- AbsenceCaptured Jul 17, 2026
No public incident/status page, model-update changelog, or continuous AI-monitoring dashboard was found.
Customer Documentation
4 items60
- DocumentationCaptured Jul 17, 2026
Detailed compliance guidance explains EEOC/Uniform Guidelines expectations, validation, and cut-scores, and notes PI's EFPA certification for deployers.
https://www.predictiveindex.com/learn/support/legality-and-compliance-of-pi-assessments/
- DocumentationCaptured Jul 17, 2026
Public product and validity pages describe the behavioral and cognitive assessments and their evidence base in depth.
https://www.predictiveindex.com/assessments/assessments-validity/
- DocumentationCaptured Jul 17, 2026
A privacy hub links website/services privacy policies, the Data Privacy Framework page, and a privacy-and-DPA FAQ for customers.
- AbsenceCaptured Jul 17, 2026
No NYC Local Law 144 candidate-notice/compliance guidance or EU AI Act deployer documentation was found in PI's resources.
Editorial notes
Company overview
The Predictive Index (Predictive Index, LLC) is a talent-optimization company founded in 1955 by Arnold Daniels and headquartered in Westwood, Massachusetts; it celebrated its 70th anniversary in February 2025 and was acquired from the Daniels family by Mike Zani and Daniel Muzquiz in 2014. Its core hiring products are the forced-choice PI Behavioral Assessment and the PI Cognitive Assessment, backed by 400+ criterion-validity studies, ongoing differential item functioning (DIF) fairness analyses, and third-party EFPA psychometric certification. In 2025-2026 PI layered a generative-AI assistant ('Obi') on top of the platform for tasks like interview-feedback synthesis, meeting notes, reviews, and job-ad generation, positioned as a support tool rather than a decision-maker.
Regulatory exposure
Used in hiring, PI's assessments are the kind of scored selection tool that NYC Local Law 144 treats as an AEDT and the EU AI Act (Annex III) treats as high-risk, yet PI publishes none of the regulation-specific artifacts: there is no public NYC LL 144 bias/adverse-impact audit, no EU AI Act Fundamental Rights Impact Assessment or deployer guidance, no AI system/model cards, and no ISO 42001. What PI does publish is strong traditional psychometric evidence (public validity and DIF methodology pages, EFPA third-party certification, and EEOC/Uniform Guidelines compliance guidance) plus a data-security posture (Vanta Trust Center, SOC 2 and GDPR claims, EU-US/UK/Swiss Data Privacy Framework certification, DPA on request). The result is a vendor that is well-documented on assessment science and data governance but leaves AI-Act/LL-144-specific compliance largely to the deploying employer.
Path to a higher score
PI could raise its score most by commissioning and publicly posting a downloadable NYC LL 144-style independent bias/adverse-impact audit (e.g., BABL AI, Warden AI, DCI) and repeating it annually, and by publishing EU AI Act deployer materials including a FRIA template and Article 26 obligation guidance. It should convert its gated technical validity summary and EFPA certificate into public downloads, produce AI/model documentation (system cards, instructions-for-use) for both the assessments and the Obi assistant, pursue ISO 42001, and document concrete human-oversight controls (override, audit logs, per-jurisdiction settings) rather than general 'decision-support' framing.
Regulatory frame
What applies to assessment AI.
Assessment is the most-regulated surface in AI hiring. NYC LL 144, EEOC and OFCCP frameworks, Illinois HB 3773, and EU AI Act Annex III §4 all apply, with overlap on bias-audit obligations.
Compare
The Predictive Index against its nearest-scoring peers.
The closest scores in Assessment AI, side by side.
- Side-by-sideThe Predictive Index vs X0PA AI48F
- Side-by-sideThe Predictive Index vs Sova Assessment48F
- Side-by-sideThe Predictive Index vs VidCruiter48F
- Side-by-sideThe Predictive Index vs Saville Assessment48F
Ranking
Others rated in Assessment AI
Ranked 29 of 82 by weighted total under rubric v1.0. The ordering is arithmetic on the rubric and carries no view on which tool suits a given hiring process.
- #26X0PA AI48F
- #27Sova Assessment48F
- #28VidCruiter48F
- #29The Predictive IndexThis profile48F
- #30Saville Assessment48F
- #31Thomas International48F
Conflicts of interest
No vendor pays for placement, scoring, or removal. Casework - the consulting firm that operates this directory - provides paid services to some vendors. Any active or recent (within 24 months) commercial relationship is disclosed on the affected vendor profile and the review is reassigned to an independent reviewer. See the full policy on About.
Casework has no commercial relationship with this vendor.