SP
BravenNow
Evaluating the Reliability of Digital Forensic Evidence Discovered by Large Language Model: A Case Study
| USA | technology | ✓ Verified - arxiv.org

Evaluating the Reliability of Digital Forensic Evidence Discovered by Large Language Model: A Case Study

#Digital Forensics #Large Language Models #AI Evidence Reliability #Forensic Investigation #Knowledge Graph #Chain of Custody #Artifact Extraction #Cybersecurity

📌 Key Takeaways

  • Researchers developed a framework to evaluate AI-identified digital evidence reliability
  • The framework automates artifact extraction and uses LLM-driven analysis with validation through a Digital Forensic Knowledge Graph
  • Tested on a 13 GB dataset with 61 applications, 2,864 databases, and 5,870 tables
  • Achieved over 95% accuracy in artifact extraction with strong chain-of-custody adherence

📖 Full Retelling

Researchers Jeel Piyushkumar Khatiwala, Daniel Kwaku Ntiamoah Addai, and Weifeng Xu have developed a structured framework to evaluate the reliability of digital evidence discovered by large language models at the IEEE 49th Annual Computers, Software, and Applications Conference in Toronto, Canada on February 22, 2026, addressing growing concerns about AI-identified evidence in forensic investigations. The research introduces an innovative approach that automates forensic artifact extraction, refines data through LLM-driven analysis, and validates results using a Digital Forensic Knowledge Graph (DFKG). The framework was tested on a substantial 13 GB forensic image dataset containing 61 applications, 2,864 databases, and 5,870 tables, ensuring artifact traceability and evidentiary consistency through deterministic Unique Identifiers and forensic cross-referencing. According to the study, this methodology addresses critical challenges in maintaining the credibility and forensic integrity of AI-identified evidence, significantly reducing classification errors while advancing scalable, auditable methodologies that could transform how law enforcement and cybersecurity professionals handle digital evidence in court proceedings.

🏷️ Themes

Digital Forensics, Artificial Intelligence, Evidence Reliability

📚 Related People & Topics

Forensic science

Forensic science

Application of science to criminal and civil laws

Forensic science, often confused with criminalistics, is the application of science principles and methods to support decision-making related to rules or law, generally criminal and civil law. During criminal investigation in particular, it is governed by the legal standards of admissible evidence a...

View Profile → Wikipedia ↗

Knowledge Graph

Topics referred to by the same term

A knowledge graph is a knowledge base that uses a graph-structured data model.

View Profile → Wikipedia ↗

Large language model

Type of machine learning model

A large language model (LLM) is a language model trained with self-supervised machine learning on a vast amount of text, designed for natural language processing tasks, especially language generation. The largest and most capable LLMs are generative pre-trained transformers (GPTs) that provide the c...

View Profile → Wikipedia ↗

Entity Intersection Graph

Connections for Forensic science:

👤 The Search 1 shared
🌐 Disappearance of Nancy Guthrie 1 shared
🌐 Glove 1 shared
🌐 Investigation 1 shared
🌐 DNA profiling 1 shared
View full profile
Original Source
--> Computer Science > Cryptography and Security arXiv:2602.20202 [Submitted on 22 Feb 2026] Title: Evaluating the Reliability of Digital Forensic Evidence Discovered by Large Language Model: A Case Study Authors: Jeel Piyushkumar Khatiwala , Daniel Kwaku Ntiamoah Addai , Weifeng Xu View a PDF of the paper titled Evaluating the Reliability of Digital Forensic Evidence Discovered by Large Language Model: A Case Study, by Jeel Piyushkumar Khatiwala and 2 other authors View PDF HTML Abstract: The growing reliance on AI-identified digital evidence raises significant concerns about its reliability, particularly as large language models are increasingly integrated into forensic investigations. This paper proposes a structured framework that automates forensic artifact extraction, refines data through LLM-driven analysis, and validates results using a Digital Forensic Knowledge Graph . Evaluated on a 13 GB forensic image dataset containing 61 applications, 2,864 databases, and 5,870 tables, the framework ensures artifact traceability and evidentiary consistency through deterministic Unique Identifiers and forensic cross-referencing. We propose this methodology to address challenges in ensuring the credibility and forensic integrity of AI-identified evidence, reducing classification errors, and advancing scalable, auditable methodologies. A comprehensive case study on this dataset demonstrates the framework's effectiveness, achieving over 95 percent accuracy in artifact extraction, strong support of chain-of-custody adherence, and robust contextual consistency in forensic relationships. Key results validate the framework's ability to enhance reliability, reduce errors, and establish a legally sound paradigm for AI-assisted digital forensics. Comments: 10 pages, 5 figures. Published in the Proceedings of the 2025 IEEE 49th Annual Computers, Software, and Applications Conference , Toronto, ON, Canada, 8-11 July 2025 Subjects: Cryptography and Security (cs.CR) ; Artificial Int...
Read full article at source

Source

arxiv.org

More from USA

News from Other Countries

🇬🇧 United Kingdom

🇺🇦 Ukraine