Qureos
Qureos is a Dubai-based AI recruiting platform whose Iris agents source candidates, score and rank applicants against recruiter-set criteria, and run AI voice and video interviews with personality scoring, alongside an ATS and a job-seeker marketplace focused on MENA and South Asia.
HireAIScore rates Qureos 33 out of 100 (grade F, Substantial gaps) under rubric v1.0, last reviewed September 25, 2026. Qureos ranks 151 of 160 vendors rated in Screening AI, against a category average of 46. That score is drawn from 35 evidence items across 7 rubric criteria for Qureos — 28 with a cited source, 7 recording that nothing was located.
§ 01 - Company facts
- Legal name
- Qureos Inc.
- Founded
- 2021
- Headquarters
- AE
- Pricing tier
- Mid
- Market side
- Employer-side (AEDT)
- Categories
- ScreeningSourcingVideo interview
- Website
- qureos.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%28
+5.6 · category avg 49
- 02
Bias Audit Transparency
Weight 18%12
+2.2 · category avg 41
- 03
FRIA Support
Weight 15%25
+3.8 · category avg 31
- 04
Data Governance Disclosure
Weight 15%44
+6.6 · category avg 53
- 05
Human Oversight Design
Weight 12%54
+6.5 · category avg 56
- 06
Post-Market Monitoring
Weight 12%35
+4.2 · category avg 40
- 07
Customer Documentation
Weight 8%52
+4.2 · category avg 58
Category avg is the mean raw score on that criterion across the 160 Screening AI vendors in scope of this rubric, this one included.
§ 03 - Strongest · weakest
Strongest category
Human Oversight Design
Raw score 54 · contributes 6.5 to total.
54 against a 56 category average
Weakest category
Bias Audit Transparency
Raw score 12 · contributes 2.2 to total.
12 against a 41 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
- 35
- Documentation
- 19
- Public statement
- 9
- Absence
- 7
- With a source URL
- 28 of 35
- Source hostnames
- 3
- Fewest items
- 2
- FRIA Support
These figures measure how thoroughly Qureos was reviewed, not how Qureos performed — an absence row, recording that nothing was located, is counted like any other item.
Article 11 Technical Documentation
5 items28
- AbsenceCaptured Sep 25, 2026
No model card, technical documentation, explainability statement, AI principles or ethics page, or ISO 42001 claim was found; qureos.com/responsible-ai, /ai-ethics, /ethics, /trust, /security and /compliance all return 404 and the site's sitemap lists no such page.
- Public statementCaptured Sep 25, 2026
The only public description of interview scoring is a marketing guide (last updated 6 July 2026) saying the AI combines NLP, sentiment analysis of 'emotional tone' and assessment of 'facial expressions and body language' into a 'comprehensive score', with no validation, error-rate or limitation data.
https://www.qureos.com/hiring-guide/guide-to-ai-video-interviews-for-employers
- Public statementCaptured Sep 25, 2026
The June 2026 Personality Analysis release says each selected OCEAN trait gets a score, confidence level and behavioural summary and calls the framework 'validated', but links no validation study, norms or accuracy figures.
- DocumentationCaptured Sep 25, 2026
The public knowledge base explains how recruiters configure AI Screening criteria and weight personality traits and interview answers into an overall score, which is operating instructions rather than technical documentation of the models.
- DocumentationCaptured Sep 25, 2026
The DPA's subprocessor table is the only disclosure of the model supply chain, naming OpenAI (NLP, embeddings, speech-to-text), Google Gemini (candidate profile analysis) and Deepgram (speech-to-text and audio analysis).
Bias Audit Transparency
7 items12
- AbsenceCaptured Sep 25, 2026
No NYC LL 144 bias audit, independent fairness audit or academic audit is published or referenced; trust.warden-ai.com/qureos returns 404, trust.app.holisticai.io/qureos returns only a generic app shell, and qureos.com has no audit, fairness or responsible-AI page.
- Public statementCaptured Sep 25, 2026
The March 2024 AI Video Interview launch post claims 'unbiased behavioral profiles' of fit 'regardless of gender, age, or ethnicity' with no audit, sample or method behind it.
- DocumentationCaptured Sep 25, 2026
The knowledge base documents sourcing filters that let recruiters pick a gender ('Both, Male, or Female'), specify a nationality, and 'Exclude specific nationalities', and no fairness testing of these features is published.
- Public statementCaptured Sep 25, 2026
Qureos's 2025 AI Screening announcement lists 'nationality, gender' among the criteria hiring teams can define for the AI to score and rank profiles against.
https://www.qureos.com/updates/ai-screening-to-streamline-candidate-evaluation
- Public statementCaptured Sep 25, 2026
The nationality-filter update says the system makes 'an educated guess about nationality, getting it right about 9 times out of 10' and pitches it for nationalisation quotas, with no error or fairness analysis.
https://www.qureos.com/updates/new-feature-update-filter-candidates-by-nationality
- Public statementCaptured Sep 25, 2026
The talent screening page promises 'bias-free shortlisting' while stating 'Iris adds details like nationality, gender, work history, and key skills to every profile.'
- Public statementCaptured Sep 25, 2026
Qureos's own March 2026 guide tells buyers to ask vendors whether their model has undergone a disparate-impact audit, yet Qureos publishes no such audit of its own tools.
https://www.qureos.com/hiring-guide/best-ai-recruiting-platforms
FRIA Support
2 items25
- AbsenceCaptured Sep 25, 2026
No EU AI Act deployer guidance, instructions for use, FRIA template or DPIA support was found on qureos.com (product pages, knowledge base, Updates log, hiring guides) or in the app policy hub (candidate and company privacy policies and terms, cookie policy, DPA).
- DocumentationCaptured Sep 25, 2026
The DPA covers GDPR processor duties (subprocessor notice, SCCs, help with data-subject requests, breach notification, customer audit rights) but says nothing about fundamental-rights impact assessments or AI Act deployer obligations.
Data Governance Disclosure
7 items44
- DocumentationCaptured Sep 25, 2026
The public DPA (updated 15 August 2025) lists purposes including AI-assisted screening and interview scoring, data categories including interview audio, video and transcripts, and eight named subprocessors with purposes and certifications, and says Qureos will not use company personal data to train or fine-tune AI models.
- DocumentationCaptured Sep 25, 2026
The candidate privacy policy (effective 2 May 2025) sets region-specific storage (EU, UAE, KSA), deletes data after 12 months of inactivity or within 30 days of a request, and lists nationality and gender only as optional fields, without mentioning AI inference, automated decision-making or emotion analysis.
- DocumentationCaptured Sep 25, 2026
The company privacy policy says AI is used for 'candidate profile analysis, job description generation, and interview transcription', repeats the no-training commitment, retains data for 90 days after termination, and cites ISO and SOC certifications held by subprocessors rather than Qureos.
- Public statementCaptured Sep 25, 2026
A September 2024 product update says 'Our AI now provides accurate details about your candidates, including their gender', which the candidate privacy policy does not disclose as an AI inference.
https://www.qureos.com/updates/advanced-candidate-management-meets-smart-data-enrichment
- DocumentationCaptured Sep 25, 2026
The candidate terms say recordings are 'not used for biometric identification' and are 'evaluated based on professional criteria', which does not address the emotional-cue and facial-expression analysis described on Qureos's product pages.
- DocumentationCaptured Sep 25, 2026
The privacy policy linked from the qureos.com footer is a generic template naming Qureos, Inc. at a Delaware address ('Last Published - 11/02/2026') and never mentions AI, interview recordings, biometric data or specific retention periods.
- AbsenceCaptured Sep 25, 2026
No Qureos-held SOC 2 report or ISO 27001/42001 certificate was found: trust.qureos.com and qureos.safebase.io do not resolve, app.vanta.com/qureos/trust returns the same generic shell as a nonexistent slug, and qureos.com/security returns 404.
Human Oversight Design
5 items54
- DocumentationCaptured Sep 25, 2026
The homepage FAQ says every screening decision 'comes with a clear rationale' that recruiters can 'override it any time', and the agents section says 'You stay in control of every decision.'
- DocumentationCaptured Sep 25, 2026
The knowledge base describes AI Screening results as per-criterion Good match, Potential fit or Not a match labels with AI-generated explanations on a detailed evaluation page, plus recruiter-ranked criteria and thumbs-up/down ratings.
- DocumentationCaptured Sep 25, 2026
Personality Analysis returns a score, a confidence level and a behavioural summary 'grounded in what the candidate actually said' for each recruiter-selected trait.
- Public statementCaptured Sep 25, 2026
The AI video interview page says the software 'ensures that the most qualified candidates proceed' and that reports combine emotional cues, a feeling chart, personality ratings and an overall score, which amounts to automated gating with no documented human-review requirement.
- AbsenceCaptured Sep 25, 2026
No per-jurisdiction controls (such as switching off emotion or facial analysis in the EU or running Illinois AI Video Interview Act consent flows), no candidate-facing explanation or appeal route, and no decision audit-log documentation were found in the knowledge base, product pages or policies.
Post-Market Monitoring
4 items35
- DocumentationCaptured Sep 25, 2026
An unlinked Better Stack status page monitors three endpoints (app.qureos.com, apiv3aws.qureos.com, places.qureos.com/healthcheck) with 90-day uptime bars, and its incident history reports no incidents from July to September 2026.
- DocumentationCaptured Sep 25, 2026
The public Updates log records feature launches from December 2023 to June 2026, such as AI Screening and Personality Analysis, but no model updates, versioning, or accuracy and bias re-testing.
- DocumentationCaptured Sep 25, 2026
The DPA commits to audit logging and monitoring, regular vulnerability scanning and breach notification 'without undue delay', with no public detail on how these are carried out.
- AbsenceCaptured Sep 25, 2026
No security or vulnerability-disclosure contact was found: qureos.com/.well-known/security.txt and /security return 404, the only contacts found were hello@, legal@, support@ and customer.success@qureos.com, and there is no public monitoring of AI performance or fairness.
Customer Documentation
5 items52
- DocumentationCaptured Sep 25, 2026
A public Iris knowledge base walks recruiters through sourcing filters, AI Screening, job pipelines, talent pools and video assessments.
- DocumentationCaptured Sep 25, 2026
An app policy hub publishes separate candidate and company privacy policies and terms, a cookie policy, and a Data Processing Agreement with a named subprocessor list.
- DocumentationCaptured Sep 25, 2026
Product pages such as the AI video interview page describe features and outputs (question library, Big Five traits, reports with an overall score) but carry no legal caveats or guidance on lawful use.
- DocumentationCaptured Sep 25, 2026
The only regulatory content is a generic section in a listicle that ranks Qureos first, summarising EEOC guidance, NYC LL 144 and the Illinois AI Video Interview Act without explaining how Qureos helps deployers comply.
https://www.qureos.com/hiring-guide/best-ai-recruiting-platforms
- AbsenceCaptured Sep 25, 2026
No ethics or responsible-AI page, no Qureos-specific NYC LL 144, Illinois AIVI or EU AI Act deployer guidance, and no candidate AI-notice templates were found on qureos.com, in the knowledge base or in the app policy hub.
§ 05 - Editorial notes
Company overview
Qureos was founded in Dubai in 2021 by Alexander Epure, Usama Nini and Mehrad Yaghmai as a career-matching platform. It now sells Iris, a set of AI recruiting agents that source candidates, screen and rank applicants against recruiter-set criteria, run AI voice and video interviews with Big Five personality scoring, and handle scheduling and outreach. It also offers an ATS and a free job-seeker marketplace. Its focus is the GCC, the wider MENA region and South Asia, and it says it is expanding into Europe and the US. Named customers include Qatar Airways, Dubai Economy and Tourism and DP World. In February 2026 it announced a $5M seed round led by Prosus Ventures and Salica Oryx Fund. The contracting entity is Qureos Inc.
Regulatory exposure
AI Screening scores and ranks applicants, and AI interviews produce an overall candidate score. Both are high-risk employment uses under EU AI Act Annex III (obligations now due 2 December 2027) and are likely AEDTs under NYC Local Law 144 wherever customers hire in NYC. The more immediate issue is Article 5(1)(f). Qureos's live AI video interview page says recruiters can analyse candidates' 'expressions, and demeanor', with reports showing 'emotional cues' and a 'feeling chart'. A Qureos guide updated in July 2026 says its AI assesses 'facial expressions and body language'. The Commission's guidelines apply the workplace emotion-recognition ban to job candidates and limit it to inferences from biometric data, which is what these descriptions suggest. That ban has applied since 2 February 2025, and Qureos publishes nothing saying the analysis is disabled for EU use. Separately, recruiters can filter by gender, include or exclude nationalities, and set nationality or gender as AI screening criteria. The tooling is built for Gulf nationalisation quotas, but applied to US candidates it invites disparate-treatment claims under Title VII, Illinois HB 3773 (in force since 1 January 2026) and California's FEHA ADS rules. The Illinois AI Video Interview Act's notice, consent and deletion duties apply to its video interviews, and Colorado's SB 26-189 disclosure duties start on 1 January 2027.
Path to a higher score
The biggest gains would come from a dated public statement on whether video interviews still infer emotions or analyse facial expressions and body language, and withdrawing that capability, at least for EU use. Next would be removing or geofencing the gender and nationality filters and screening criteria so they work only where nationalisation law requires them, then commissioning and publishing an independent LL 144-style bias audit of AI Screening and AI Interview scoring. After that, Qureos should: publish technical documentation covering scoring logic, validation of its Big Five claims, accuracy and limitations; align its privacy notices with the product, including disclosing AI inference of nationality and gender; provide candidate AI notices, Illinois AIVI consent support and EU AI Act deployer and FRIA guidance; and obtain its own SOC 2 or ISO 27001 attestation, publish a security contact and link its status page.
§ 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
Build any comparison→Qureos against its nearest-scoring peers.
In Screening AI.
§ Others rated in Screening AI
All screening vendors→Ranked 151 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.
- 148Zappyhire34F
- 149HireQuotient34F
- 150Pulsifi33F
- 151QureosThis profile33F
- 152Entelo33F
- 153Emi33F
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.