ConverzAI
HireAIScore rates ConverzAI 44 out of 100 (grade F, Substantial gaps) under rubric v1.0, last reviewed September 25, 2026. ConverzAI ranks 90 of 160 vendors rated in Screening AI, against a category average of 46. That score is drawn from 30 evidence items across 7 rubric criteria for ConverzAI — 24 with a cited source, 6 recording that nothing was located.
Score breakdown
ConverzAI, 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 49
30
+6.0
Bias Audit Transparency
Weight 18% · category avg 41
72
+13.0
FRIA Support
Weight 15% · category avg 31
30
+4.5
Data Governance Disclosure
Weight 15% · category avg 53
40
+6.0
Human Oversight Design
Weight 12% · category avg 56
45
+5.4
Post-Market Monitoring
Weight 12% · category avg 40
50
+6.0
Customer Documentation
Weight 8% · category avg 58
45
+3.6
Total weighted score
44.5(rounded to 44 for display)
Category avg is the mean raw score on that criterion across the 160 Screening AI vendors in scope of this rubric, this one included.
Strongest criterion
Bias Audit Transparency
72raw / 100
Contributes 13.0 to the total. 72 against a 41 category average.
Weakest criterion
FRIA Support
30raw / 100
Contributes 4.5 to the total. 30 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
- 30
- Documentation
- 11
- Audit report
- 6
- Public statement
- 7
- Absence
- 6
- With a source URL
- 24 of 30
- Source hostnames
- 3
- Fewest items
- 2
- FRIA Support
These figures measure how thoroughly ConverzAI was reviewed, not how ConverzAI performed — an absence row, recording that nothing was located, is counted like any other item.
Article 11 Technical Documentation
3 items30
- DocumentationCaptured Sep 25, 2026
The only public system description is on ConverzAI's Warden AI dashboard: 'AI Candidate Matching' sends job-related questions and turns job criteria and candidate responses into a match score reflecting skills, preferences, expectations and other factors.
- Public statementCaptured Sep 25, 2026
The product page says the Virtual Recruiter builds role-specific interview questions from job details (skills, role type, pay), 'makes its own decisions' and is 'backed by multiple patents', but gives no model documentation, scoring logic or instructions for use.
- AbsenceCaptured Sep 25, 2026
No model or system card, responsible-AI policy, explainability statement, instructions for use or ISO/IEC 42001 certification was found in all 28 pages and 12 posts exposed by converzai.com's WordPress REST API, its page sitemap, docs.converzai.com (which redirects to a corporate wiki host that does not resolve publicly) or support.converzai.com (sign-in required).
Bias Audit Transparency
6 items72
- Audit reportCaptured Sep 25, 2026
Warden AI's NYC LL 144 report on 'ConverzAI - AI Candidate Matching', dated 27 October 2025 (sample 18,007, with 14,288 in the sex analysis), reports sex impact ratios of 1.00 and 0.94, race/ethnicity ratios of 0.92 to 1.00, and a lowest intersectional ratio of 0.80 for Asian men, excluding groups under 2% of the sample.
- Audit reportCaptured Sep 25, 2026
The same report contradicts itself on its data: the summary says Warden's own dataset was used 'due to a lack of access to historical data' and calls the audit 'for demonstration purposes', but the audit-details table lists a bulk export of ConverzAI's historical data.
- Audit reportCaptured Sep 25, 2026
Warden's public dashboard labels the production-data matching system as audited 'Monthly', but its latest audit is still 27 October 2025 (14,288 samples; sex, race/ethnicity and intersectional results all 'Clear') and it carries only an NYC LL 144 report.
- Audit reportCaptured Sep 25, 2026
A second Warden system, run on Warden's test dataset and audited on 23 September 2026 (27,824 samples), shows 102 'Clear' results across 15 bias categories including age, disability and national origin, with reports mapped to NYC LL 144, the EU AI Act, Colorado SB 26-189 and California FEHA.
https://trust.warden-ai.com/converzai/ai-candidate-matching-test-data
- DocumentationCaptured Sep 25, 2026
ConverzAI's 'AI Bias Audit 2025' page names Warden AI and links the LL 144 report for download, but the page is missing from the site navigation and page sitemap, and the homepage's 'Third-party AI bias audited' badge does not link to it.
- Public statementCaptured Sep 25, 2026
A solution page markets 'Bias-free screening at scale', an absolute claim that goes further than the Warden reports, which state their findings do not guarantee the absence of bias.
https://www.converzai.com/solutions/enhance-candidate-experience/
FRIA Support
2 items30
- AbsenceCaptured Sep 25, 2026
No Fundamental Rights Impact Assessment template, EU AI Act deployer guidance or Article 26/27 material was found in converzai.com's pages and posts (via its WordPress REST API and sitemaps), the privacy policy, the AI Bias Audit page or support.converzai.com (sign-in required).
- DocumentationCaptured Sep 25, 2026
Warden's EU AI Act bias-audit report for the test-data system (23 September 2026) restates generic provider and deployer duties such as risk management, post-market monitoring and human oversight, and says it is not evidence of full compliance; it is a bias test, not a fundamental-rights assessment aid.
https://trust.warden-ai.com/converzai/ai-candidate-matching-test-data/eu-ai-act
Data Governance Disclosure
4 items40
- Public statementCaptured Sep 25, 2026
The homepage calls the product 'Safe, secure and 100% compliant' with SOC 2 Type II, HIPAA and third-party bias-audit badges, but the badges are unlinked and no SOC 2 report, auditor name or trust center is published.
- DocumentationCaptured Sep 25, 2026
The privacy notice says customers' policies govern candidate data, extends CCPA-style rights to all users via privacy@converzai.com, hosts data on AWS with encrypted transfers, reserves commercial use of aggregated de-identified data, and retains data 'for as long as needed' or as required by law, with no defined periods.
- DocumentationCaptured Sep 25, 2026
Warden's system summary is the only public statement of what data the matching model uses, listing job criteria and candidate responses as inputs and a match score as output.
- AbsenceCaptured Sep 25, 2026
No training-data description, excluded-attribute list, retention schedule for candidate conversation data (calls, SMS, transcripts or summaries), DPA or published subprocessor list was found in the privacy policy, the product, solution and case-study pages, all 12 blog posts, the WordPress media library or the trust-center hosts (trust.converzai.com and converzai.safebase.io do not resolve).
Human Oversight Design
5 items45
- DocumentationCaptured Sep 25, 2026
The Integrity Staffing case study says the Virtual Recruiter 'doesn't make hiring decisions or place candidates'; it surfaces a candidate's skills against the job description so the firm's recruiters decide.
- DocumentationCaptured Sep 25, 2026
ConverzAI says the Virtual Recruiter 'screens and scores candidates so recruiters receive a clean, prioritized shortlist - ready for submission', a decision-support framing with no documented review, override or score-explanation controls.
https://www.converzai.com/solutions/increase-recruiter-efficiency/
- Public statementCaptured Sep 25, 2026
Marketing stresses autonomy: the agent 'makes its own decisions' and works 'on autopilot eliminating human delays', and a TalentBurst tile reports '14% Handsfree placements'.
- Public statementCaptured Sep 25, 2026
At TalentBridge, the AI recruiter 'Caroline' initiates outreach, pre-screens candidates and updates the ATS 'all without human intervention'.
- AbsenceCaptured Sep 25, 2026
No public documentation of recruiter override or appeal paths, per-candidate score explanations, audit logs or per-jurisdiction settings was found in the product, solution and case-study pages, all blog posts or support.converzai.com (sign-in required).
Post-Market Monitoring
4 items50
- Audit reportCaptured Sep 25, 2026
Warden's public dashboard for ConverzAI advertises 'Continuous bias auditing' and 'Post-market monitoring' and shows a monthly test-data audit last run on 23 September 2026.
https://trust.warden-ai.com/converzai/ai-candidate-matching-test-data
- Audit reportCaptured Sep 25, 2026
Monitoring of the production-data system appears to have lapsed: the dashboard still shows 27 October 2025 as the latest audit despite a 'Monthly' frequency.
- DocumentationCaptured Sep 25, 2026
An unlinked Better Stack status page tracks only converzai.com and portal.converzai.com, showing 100% uptime over 90 days, no posted incidents or maintenance, and a support@converzai.com contact.
- AbsenceCaptured Sep 25, 2026
No security.txt, vulnerability-disclosure policy, security contact, incident postmortem, model-update changelog or release note was found at /.well-known/security.txt (redirects to the homepage), status.converzai.com (no DNS record), converzai.statuspage.io (unregistered), the status page's RSS feed (no entries) or any product or blog page.
Customer Documentation
6 items45
- DocumentationCaptured Sep 25, 2026
The product page covers the Virtual Recruiter's voice, SMS and email engagement, candidate insights, autonomous sourcing, smart interviews, ATS updates and performance dashboards, and shows ATS integrations including Bullhorn, Avionté and JobDiva.
- DocumentationCaptured Sep 25, 2026
An October 2025 blog post advises staffing firms to tell candidates up front that a virtual recruiter will contact them, and assigns submission and final hiring decisions to humans.
- DocumentationCaptured Sep 25, 2026
A blog post advises monitoring outcomes such as demographic diversity and offer rates, periodically auditing the AI's decisions and leaving the final go/no-go call to a recruiter.
- Public statementCaptured Sep 25, 2026
A November 2025 blog FAQ says 'Candidates often don't realize they're interacting with AI initially', which undercuts ConverzAI's own advice to disclose the virtual recruiter up front.
https://www.converzai.com/thanksgiving-ai-recruiting-automation/
- Public statementCaptured Sep 25, 2026
A candidate testimonial on a solution page, presented as praise, says 'It took me a while to realize it was a Virtual Recruiter and not a real person'.
https://www.converzai.com/solutions/enhance-candidate-experience/
- AbsenceCaptured Sep 25, 2026
No DPA, terms of service, NYC LL 144, Illinois or Colorado deployer guidance, candidate-notice template, AI-call disclosure or TCPA consent policy, or public help center was found in converzai.com's page and post sitemaps and REST API, the resource library (no compliance content), docs.converzai.com (non-public corporate wiki) or support.converzai.com (Zoho Desk sign-in).
Editorial notes
Company overview
ConverzAI Inc. is a Seattle-based company founded in 2019 by Ashwarya Poddar, who previously spent more than a decade at Microsoft, and is backed by a Series A led by Menlo Ventures (announced in February 2025). Its Virtual Recruiter is an agentic AI agent for US staffing firms. It sources from existing databases, job boards and recruiter hotlists, contacts applicants by voice, SMS and email within minutes of an application, runs role-specific screening interviews built from job details, and scores each candidate's fit to open jobs. It also schedules interviews, chases paperwork, and writes notes and summaries back to ATSs such as Bullhorn, Avionté and JobDiva. It is sold on outcome-based pricing ('pay only for placements'), and its published case studies include Integrity Staffing Solutions, TalentBridge, TalentBurst, Malone Workforce Solutions, The Advance Group, SSi People and I.K. Hofmann USA.
Regulatory exposure
Staffing firms are employment agencies under NYC Local Law 144, and a match score used to screen and rank applicants can count as an AEDT, depending on how much weight it carries. ConverzAI's customers therefore need a bias audit less than a year old, a published summary and advance notice to candidates. ConverzAI supplies a Warden AI summary, but its production-data audit dates from 27 October 2025 and the newer reports use Warden's test data. That gap matters more now that DCWP has promised tougher enforcement after the December 2025 State Comptroller audit. Illinois HB 3773 (from 1 January 2026) requires notice when AI is used in employment decisions. Colorado SB 26-189 adds disclosure duties for automated decision-making technology from 1 January 2027. California's FEHA automated-decision rules require four-year record retention, which ConverzAI's open-ended retention language does not address. Because the agent places outbound AI-voice calls, consent and disclosure are central: the FCC ruled in February 2024 that AI-generated voices are 'artificial' under the TCPA. ConverzAI publishes no call-consent or AI-disclosure policy, and its own blog says candidates 'often don't realize' they are talking to AI. No EU deployment is evidenced, but any EU use would be high-risk under Annex III point 4 of the EU AI Act, with those obligations now deferred to 2 December 2027.
Path to a higher score
ConverzAI's quickest gains come from publishing what it already has and fixing what is inconsistent. It should link the Warden dashboard and bias-audit page from its main navigation. It should re-run the production-data audit before the October 2025 report turns a year old (the system is labelled monthly) and correct the report's contradictory statements about its data. It should publish a system card for the match score covering inputs, excluded attributes, how scores feed shortlists and known limits. It should document recruiter review and override controls and say plainly whether any candidates are rejected or placed without human review, a question its 'handsfree placements' marketing raises. Customers need LL 144, Illinois and Colorado notice templates and a written AI-call disclosure and TCPA consent policy. It should also publish a DPA with a subprocessor list, a SOC 2 report or trust center, and a retention schedule for conversation data that supports customers' four-year California record-keeping. A security contact is also missing, and it needs EU deployer (FRIA) guidance if it sells outside the US.
Regulatory frame
What applies to screening AI.
Treated as high-risk under Annex III §4 when the screening output is decisive. EU AI Act Article 13 transparency obligations apply; deployers must give candidates a way to contest.
Compare
ConverzAI against its nearest-scoring peers.
The closest scores in Screening AI, side by side.
- Side-by-sideConverzAI vs Vincere44F
- Side-by-sideConverzAI vs Equalture44F
- Side-by-sideConverzAI vs Emmersion44F
- Side-by-sideConverzAI vs Fountain44F
Ranking
Others rated in Screening AI
Ranked 90 of 160 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.
- #87Avionté45F
- #88Cangrade45F
- #89Cegid45F
- #90ConverzAIThis profile44F
- #91Vincere44F
- #92Equalture44F
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.