Large Language Models and Forensic Linguistics: Navigating Opportunities and Threats in the Age of Generative AI

TL;DR

The paper assesses how large language models reshape forensic linguistics by expanding capabilities for authorship attribution and multilingual analysis while threatening idiolect-based evidence and detector reliability. It surveys theoretical foundations, detection methods, watermarking, and model-source attribution, highlighting both practical gains and admissibility challenges under Daubert/Kumho. The authors advocate hybrid human–AI workflows, explainable detection, and rigorous validation across diverse populations to preserve evidentiary credibility. They conclude that the core insight—language reveals its producer—remains valid but must adapt to complex human–machine authorship Chains in the generative AI era.

Abstract

Large language models (LLMs) present a dual challenge for forensic linguistics. They serve as powerful analytical tools enabling scalable corpus analysis and embedding-based authorship attribution, while simultaneously destabilising foundational assumptions about idiolect through style mimicry, authorship obfuscation, and the proliferation of synthetic texts. Recent stylometric research indicates that LLMs can approximate surface stylistic features yet exhibit detectable differences from human writers, a tension with significant forensic implications. However, current AI-text detection techniques, whether classifier-based, stylometric, or watermarking approaches, face substantial limitations: high false positive rates for non-native English writers and vulnerability to adversarial strategies such as homoglyph substitution. These uncertainties raise concerns under legal admissibility standards, particularly the Daubert and Kumho Tire frameworks. The article concludes that forensic linguistics requires methodological reconfiguration to remain scientifically credible and legally admissible. Proposed adaptations include hybrid human-AI workflows, explainable detection paradigms beyond binary classification, and validation regimes measuring error and bias across diverse populations. The discipline's core insight, i.e., that language reveals information about its producer, remains valid but must accommodate increasingly complex chains of human and machine authorship.