Inteligencia artificial en auditoría forense: análisis bibliométrico de las brechas globales de investigación y la posición de América Latina.

Abstract

The need to integrate artificial intelligence (AI) with forensic audit, forensic accounting, and detection of financial fraud is increasing. This research studies the global gaps of the field and the position of Latin America. The study is based on literature on professional judgment and audit data analytics (Alles, 2015; Appelbaum et al., 2017; Brown-Liburd et al., 2015; Kokina & Davenport, 2017), and on bibliometric approaches of science mapping (Gutiérrez-Salcedo et al., 2018; Moral-Muñoz et al., 2020). A bibliometric and thematic analysis was performed in 504 eligible publications indexed in Scopus between 2014 and 2026 through frequency analysis, keyword co-occurrence, international cooperation, thematic evolution and classification of AI methods and forensic applications. OpenAlex was used to complete metadata and to audit source bias without extending the principal corpus. Results reveal rapid growth since 2023 and predominance of automatic learning applied to financial statements fraud and transaction fraud. Despite its limited nature, regional production evidenced a greater proportional orientation towards public sector and contracting, and a relative impact quantified by number of citations, similar to global production. Therefore, it is concluded that the contribution of AI to forensic audit is growing rapidly and that regional indicators differ from the global ones. In addition, integration of AI depends not only on its technical sophistication, but also on its capacity to boost evidence, internal control, interpretability, auditability and professional supervision.