# Three New Papers Test Whether AI Agents Can Actually Do Digital Forensics

One benchmark measures whether agents can infer relational structure in undocumented mobile databases, another searches for personal data hidden inside binary fields, and a third checks which threat feeds arrive before attacks.

- Published: 2026-08-25T06:26:21.272Z
- Canonical: https://polylog.news/ai/2026-08-25/three-new-papers-test-whether-ai-agents-can-actually-do-digi
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv cs.CR (agentic forensics benchmark)](https://arxiv.org/abs/2608.21470), [arXiv cs.CR (PII discovery)](https://arxiv.org/abs/2608.21469), [arXiv cs.CR (OSINT effectiveness)](https://arxiv.org/abs/2608.21471)

Digital forensics is one of the few security tasks where agent claims can be graded objectively, because the ground truth sits in a database. Three preprints posted to arXiv today try to do exactly that. The first introduces a reproducible…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-08-25/three-new-papers-test-whether-ai-agents-can-actually-do-digi (subscription information: https://polylog.news/pricing).