Detecting Deceptive Recruitment: A Signal-theoretic Machine Learning Framework for Early Identification of Labour Exploitation

arXiv:2609.20336v1 Announce Type: new Abstract: Deceptive online job advertisements have emerged as a primary pathway into forced labour, yet systematic detection methods remain underdeveloped due to data scarcity and absence of empirically validated indicators. We formalise this detection challenge as a classification problem under signalling theory, where exploiters transmit costless signals mimicking legitimate communications across textual, visual, and structural dimensions. Using 464 verified cases (164 deceptive, 300 legitimate) collected through anti-slavery charities across nine origin countries and 21 industries, we develop multimodal detection models combining computer vision, natural language processing, and semantic embeddings. Through systematic feature ablation experiments and repeated stratified cross-validation, we demonstrate that individual modalities achieve substantial discriminatory power (ROC-AUC: 0.87--0.97), whilst their integration yields modest further gains. SHAP-based analysis reveals that text quality and domain-specific risk language are the primary discriminators, with readability indices, risk keyword density, and visa sponsorship mentions ranking highest, followed by visual colour and texture features. These production quality gaps reflect resource constraints that prevent exploiters from maintaining professional standards across all communication channels simultaneously. We operationalise findings through a proof-of-concept decision support system providing interpretable risk scores for practitioners. This work demonstrates how rigorous ana

Deceptive job ads lead to forced labour, a negative implication.

The work improves early detection of deceptive job ads leading to forced labour by integrating multiple AI modalities, which can reduce exploitation and improve human welfare. While the societal impact is positive, the score is moderate due to dependence on data quality and implementation. The approach does not imply extreme transformative effects but represents material improvement in combating labor exploitation.

This is a research proof-of-concept applying ML to detect deceptive job ads tied to forced labour, with solid AUC results but a small dataset and no deployment evidence. The implication is modestly positive: AI used defensively for a humanitarian problem, though limited scale and generalisability keep confidence in real-world impact moderate.

The development of a machine learning framework capable of detecting deceptive job advertisements is a significant positive. This innovation directly addresses a critical humanitarian challenge by helping identify and prevent forced labor, leveraging AI for societal protection.

SOURCE     arXiv cs.CY (Computers and Society) (tier 1)
   ↓
DOCUMENT   6d154538-c7be-4901-aab5-93c5d1a2cd40
           https://arxiv.org/abs/2609.20336
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  4 runs, methodology 1.0.1
   ↓
SCORE      -25  (Adverse)
   ↓
CONFIDENCE 79%