isolved
isolved is a US HCM and payroll provider whose isolved Talent Acquisition applicant tracking system (formerly ApplicantPro) uses AI Candidate Matching to score each applicant's resume against the job description and sort the applicant list, alongside AI job-ad writing and WedgeHR-powered one-way video interviews.
HireAIScore rates isolved 47 out of 100 (grade F, Substantial gaps) under rubric v1.0, last reviewed September 25, 2026. isolved ranks 34 of 77 vendors rated in ATS-integrated AI, against a category average of 47. That score is drawn from 36 evidence items across 7 rubric criteria for isolved — 31 with a cited source, 5 recording that nothing was located.
§ 01 - Company facts
- Legal name
- isolved, Inc.
- Founded
- 1986
- Headquarters
- United States · US
- Pricing tier
- Mid
- Market side
- Employer-side (AEDT)
- Categories
- ATS-integratedScreening
- Website
- isolvedhcm.com
These are company details, not findings. They are editable by a verified representative and carry no weight in the score, which is built only from the cited evidence below.
§ 02 - Score breakdown
§ Score breakdown
Category scoring
Weighted contribution shown to the right of each bar.
- 01
Article 11 Technical Documentation
Weight 20%50
+10.0 · category avg 51
- 02
Bias Audit Transparency
Weight 18%45
+8.1 · category avg 41
- 03
FRIA Support
Weight 15%30
+4.5 · category avg 32
- 04
Data Governance Disclosure
Weight 15%53
+7.9 · category avg 54
- 05
Human Oversight Design
Weight 12%55
+6.6 · category avg 57
- 06
Post-Market Monitoring
Weight 12%47
+5.6 · category avg 42
- 07
Customer Documentation
Weight 8%56
+4.5 · category avg 59
Category avg is the mean raw score on that criterion across the 77 ATS-integrated AI vendors in scope of this rubric, this one included.
§ 03 - Strongest · weakest
Strongest category
Customer Documentation
Raw score 56 · contributes 4.5 to total.
56 against a 59 category average
Weakest category
FRIA Support
Raw score 30 · contributes 4.5 to total.
30 against a 32 category average
§ 04 - Cited evidence
Download diligence record→§ Evidence
Cited per category
Every score is backed by at least one cited piece of evidence.
Evidence ledger
- Items
- 36
- Documentation
- 24
- Audit report
- 2
- Public statement
- 4
- Product feature
- 1
- Absence
- 5
- With a source URL
- 31 of 36
- Source hostnames
- 7
- Fewest items
- 4
- Article 11 Technical Documentation
These figures measure how thoroughly isolved was reviewed, not how isolved performed — an absence row, recording that nothing was located, is counted like any other item.
Article 11 Technical Documentation
4 items50
- DocumentationCaptured Sep 25, 2026
The Trust Center AI Ethics page sets out isolved's AI principles and an FAQ naming the models behind each feature, stating that Candidate Matching is built with AdeptID models and the Job Ad Writer with an OpenAI model.
- DocumentationCaptured Sep 25, 2026
The AI Use, Transparency, and Acceptable Use Policy (effective 1 August 2026) commits to risk-based classification of AI systems and to explaining functionality, key factors and known limitations where appropriate or legally required, but keeps design documentation and evaluation metrics internal.
https://www.isolvedhcm.com/legal/agreements/ai-policy-terms-of-use
- DocumentationCaptured Sep 25, 2026
The upstream AdeptID Modeling FAQs linked from isolved's Trust Center explain how the match score ensembles work-history, skills, seniority and education assessments and list its evaluation metrics; this is the model vendor's document, not isolved's.
- AbsenceCaptured Sep 25, 2026
No isolved model card, instructions for use or ISO/IEC 42001 certification was found for Candidate Matching, and there is no technical description of the video 'AI interviewer' anywhere; searched the applicantpro.com AI tools, glossary, video pages and sitemap, and the isolvedhcm.com Trust Center and legal agreements.
Bias Audit Transparency
5 items45
- Public statementCaptured Sep 25, 2026
isolved says it audits, tests and monitors for bias and that its AI Executive Committee runs audits and works with third parties to evaluate third-party AI models, but it names no auditor and publishes no results for Candidate Matching.
- Audit reportCaptured Sep 25, 2026
AdeptID, the model provider behind Candidate Matching, publishes a Credo AI 'NYC LL-144 Report Review' of its own core matching model, which is not an audit of isolved's deployment.
- Audit reportCaptured Sep 25, 2026
The downloadable review is dated 26 August 2024 on its pages, says it 'is not an independent review', relies on licensed test data, and reports impact ratios above the four-fifths rule for 2 sex, 4 race/ethnicity and 3 intersectional subgroups.
- Public statementCaptured Sep 25, 2026
AdeptID's public trust center lists 'AdeptID NYC Local Law 144 - Credo AI Assessment Report -2026.pdf' as its current NYC Local Law 144 report on its core model; the report's contents could not be opened for this assessment.
- AbsenceCaptured Sep 25, 2026
No NYC LL 144 bias audit summary for isolved Candidate Matching or its video interview AI was found on applicantpro.com or isolvedhcm.com, and Warden AI's trust directory returns 404 for both isolved and applicantpro.
FRIA Support
4 items30
- DocumentationCaptured Sep 25, 2026
The AI policy says that, subject to applicable agreements, isolved may give customers information reasonably necessary to support their compliance, audit or risk assessment requirements for AI features.
https://www.isolvedhcm.com/legal/agreements/ai-policy-terms-of-use
- Public statementCaptured Sep 25, 2026
isolved says its internal teams conduct impact assessments for new AI features, but it publishes none and offers no assessment template to customers.
- DocumentationCaptured Sep 25, 2026
The Trust Center offers additional security and audit information under NDA, with on-site reviews in some circumstances.
- AbsenceCaptured Sep 25, 2026
No EU AI Act Article 27 fundamental rights impact assessment template or EU deployer guidance was found; searched the isolvedhcm.com Trust Center, legal agreements, blog and resources sitemaps, and applicantpro.com.
Data Governance Disclosure
6 items53
- DocumentationCaptured Sep 25, 2026
isolved states that Candidate Matching's models are trained on anonymized data from millions of applicants across companies, industries and regions, that PII, race and gender are not part of training, and that PII does not influence the match score.
- DocumentationCaptured Sep 25, 2026
The AI policy says customer data, inputs and outputs are not used to train or fine-tune AI models unless the customer's contract permits it, and commits to data minimization and AI-related retention controls.
https://www.isolvedhcm.com/legal/agreements/ai-policy-terms-of-use
- DocumentationCaptured Sep 25, 2026
The Trust Center compliance page says isolved's controls are subject to annual SOC audits, without naming the report type or publishing it.
- DocumentationCaptured Sep 25, 2026
The public customer Terms and Conditions attach an isolved Data Security Addendum (Exhibit C) governing protection of the employer's Personal Information.
https://www.isolvedhcm.com/legal/agreements/terms-and-conditions
- DocumentationCaptured Sep 25, 2026
The upstream AdeptID FAQ describes its training data (including 60M resumes and profiles and 36M job postings) and says its models never see explicit demographic information or PII.
- DocumentationCaptured Sep 25, 2026
The candidate-facing isolved Talent Acquisition privacy policy (last updated 9 July 2024) does not disclose AI scoring of applicants or AdeptID's role, and its AI section covers only staff use of generative AI.
Human Oversight Design
5 items55
- Public statementCaptured Sep 25, 2026
isolved states that critical decisions, particularly in hiring, require final human review and approval before action is taken, and that users can always override or review inaccurate or biased recommendations.
- DocumentationCaptured Sep 25, 2026
The AI policy lets customers review, edit or override AI outputs, request human assistance, and escalate or challenge AI-generated outcomes, and bars customers from automating employment decisions without appropriate human oversight.
https://www.isolvedhcm.com/legal/agreements/ai-policy-terms-of-use
- DocumentationCaptured Sep 25, 2026
Candidate Matching's documented output is a match score used to sort the applicant list, and no automatic rejection is described.
https://www.applicantpro.com/glossary/ai-powered-candidate-matching
- DocumentationCaptured Sep 25, 2026
HR leaders and hiring managers review, rate and share the pre-recorded video interviews inside the ATS.
- Product featureCaptured Sep 25, 2026
WedgeHR says isolved's video interviewing is completely powered by WedgeHR and can automatically invite candidates who meet preset criteria, at application or later.
Post-Market Monitoring
6 items47
- DocumentationCaptured Sep 25, 2026
isolved runs a public Atlassian Statuspage with an Applicant Tracking component and a published incident history, although no incident carries a postmortem.
- DocumentationCaptured Sep 25, 2026
A 4 February 2026 incident notice disclosed that an ATS resume-parsing outage may impact candidate matching for resumes submitted during the window.
- DocumentationCaptured Sep 25, 2026
A 5 May 2026 incident notice disclosed an outage of the Job Ad AI Pro feature.
- DocumentationCaptured Sep 25, 2026
The AI policy commits to monitoring AI for accuracy, bias and unintended outcomes, to processes for detecting and remediating AI incidents, and to notifying customers of material incidents where law or contract requires.
https://www.isolvedhcm.com/legal/agreements/ai-policy-terms-of-use
- DocumentationCaptured Sep 25, 2026
The technical security page documents an incident response plan, annual independent security testing, audit logs kept for at least one year, and customer registration for incident updates.
- AbsenceCaptured Sep 25, 2026
No public model-update changelog, bias-monitoring results or security.txt was found for isolved Talent Acquisition (/.well-known/security.txt returns 404 on isolvedhcm.com and applicantpro.com).
Customer Documentation
6 items56
- DocumentationCaptured Sep 25, 2026
The AI hiring tools page describes AI Candidate Matching sorting applicants by score, Job Ad AI Pro, and an 'AI interviewer' guiding video interviews, with FAQs but no compliance guidance.
- DocumentationCaptured Sep 25, 2026
A Candidate Matching FAQ explains that the score reflects resume skills and experience against the job description, that thin resumes may score lower, and that the tool is for Premium clients.
https://www.applicantpro.com/glossary/ai-powered-candidate-matching
- DocumentationCaptured Sep 25, 2026
The Trust Center AI Ethics page answers governance FAQs, names model providers, and tells customers to raise AI concerns through their Customer Success Manager or a Trust Center request.
- DocumentationCaptured Sep 25, 2026
The public customer Terms and Conditions include an AI clause (section 5.6) making the employer responsible for reviewing outputs, plus the isolved Data Security Addendum as Exhibit C.
https://www.isolvedhcm.com/legal/agreements/terms-and-conditions
- DocumentationCaptured Sep 25, 2026
isolved's video interview page says the software is powered by Wedge and cites AI-driven assessments and real-time transcriptions, without describing what the AI evaluates.
https://www.isolvedhcm.com/talent-acquisition/applicant-tracking-system/video-interviews
- AbsenceCaptured Sep 25, 2026
No deployer guidance was found on NYC LL 144 notices or audits, Illinois AI Video Interview Act consent, Illinois HB 3773 notices, Colorado or California ADMT rules, or the EU AI Act; searched the applicantpro.com and isolvedhcm.com sitemaps, Trust Center, legal agreements and 'Compliance Corner' AI posts.
§ 05 - Editorial notes
Company overview
isolved (isolved, Inc.), headquartered in Charlotte, North Carolina, began in 1986 as a COBRA benefits provider and now sells the isolved People Cloud HCM, payroll and benefits platform to US employers. Its hiring product, isolved Talent Acquisition, is the former ApplicantPro (founded 2007), which isolved owned for nearly a decade before fully integrating and rebranding it in March 2025; it is sold standalone or inside People Cloud. This profile scores Candidate Matching, a Premium feature that scores each resume against the job description and sorts the applicant list (isolved's Trust Center says it is built on AdeptID models, and isolved said in October 2024 it had scored over six million applicants in nine months), and the one-way video interviews isolved markets with an 'AI interviewer', which WedgeHR says it powers entirely. Indeed Talent Scout, which runs inside the ATS, is Indeed's product and is not credited here.
Regulatory exposure
Candidate Matching gives every applicant a score that orders the list, so NYC employers who rely on it can fall within Local Law 144's AEDT definition and need an annual independent bias audit and candidate notice. isolved publishes no audit of its own. The audits in its disclosure chain cover AdeptID's upstream core model, not isolved's deployment: a Credo AI review that calls itself not independent and used licensed test data, and a 2026 Credo AI assessment report that AdeptID's trust center lists publicly but whose contents could not be reviewed for this assessment. That is a weak position now that DCWP has promised tougher enforcement after the December 2025 State Comptroller audit. In Illinois, HB 3773 (from 1 January 2026) requires notice when AI is used in employment decisions, and the AI Video Interview Act requires notice, explanation and consent before AI analyzes recorded interviews; isolved does not say what its 'AI interviewer' analyzes, while WedgeHR markets AI interview highlights and an AI feature that surfaces the strongest recorded answers first. Colorado's SB 26-189 disclosure regime starts 1 January 2027, California's FEHA automated-decision rules (with four-year record retention) have applied since 1 October 2025, and CPPA ADMT rules follow on 1 January 2027. isolved sells mainly to US employers, so EU AI Act exposure looks limited, though recruitment AI is Annex III high-risk with obligations deferred to 2 December 2027.
Path to a higher score
The biggest gain would be an independent LL 144 bias audit of Candidate Matching as deployed in isolved's ATS, refreshed yearly with a public summary, instead of relying on AdeptID's upstream reviews of its core model. isolved should also publish system cards for Candidate Matching and the video 'AI interviewer' (inputs, excluded fields, what is scored, how results are shown and overridden), add WedgeHR to its AI model disclosures, and update its July 2024 candidate privacy policy to disclose AI scoring and third-party model processing. Deployer guidance (NYC notice templates, Illinois video-interview consent and HB 3773 notices, Colorado and California ADMT duties, and an Article 27 FRIA template for any EU users), plus a public model-change log and aggregate bias-monitoring results, would lift the remaining categories.
§ Regulatory frame
What applies to ats-integrated ai.
Ranking features are inside EU AI Act Annex III §4 (high-risk) and inside NYC Local Law 144 when they affect NYC-based candidates. Customers should treat the AI surface as a distinct deployer-FRIA scope from the rest of the ATS.
§ Compare
Build any comparison→isolved against its nearest-scoring peers.
In ATS-integrated AI.
§ Others rated in ATS-integrated AI
All ats-integrated vendors→Ranked 34 of 77 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.
- 31Employment Hero48F
- 32VidCruiter48F
- 33Jobylon48F
- 34isolvedThis profile47F
- 35Teamtailor47F
- 36HiBob46F
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.