Can We Still Trust Disaster Social Sensing? Empirical Evidence on Detecting AI-Generated Social Media Posts

arXiv:2609.35821v1 Announce Type: cross Abstract: Disaster social sensing converts public social-media posts into evidence for situational awareness and humanitarian needs, but generative artificial intelligence (AI) can produce plausible messages that resemble eyewitness reports. This study investigates whether text-based AI detectors can reliably distinguish human-authored from AI-generated disaster posts. We construct a dataset of 12,000 texts organised into 3,000 matched semantic units from nine disasters: original human posts (H0), minimally LLM-proofread human posts (H1), factual AI-generated posts based on the same verified facts (A0), and affectively framed versions of those AI posts (A1). A separate 6,000-text corpus from 42 events supports model selection and threshold calibration. We evaluate OSM-Det, Fast-DetectGPT, Binoculars, and direct large language model (LLM) judges across five model families, then test disaster-domain calibration, a frozen-encoder linear readout, paired transformation sensitivity, and dataset artifact controls. Across fourteen frozen cross-family configurations, AUROC is 0.402-0.517 and the best prospective recall at a calibration-derived low-false-positive operating point is 3.6%; OSM-Det reaches AUROC 0.521 and 10.4% recall at a realised 6.7% false-positive rate. A disaster-trained linear head reaches AUROC 0.817, but a seven-feature surface classifier reaches 0.784 on the H0-versus-A0 contrast, and neutralising identified surface asymmetries reduces the head from 0.733 to 0.594. The head also separates A0 from A1 even though provenance

AI-generated posts can resemble eyewitness reports, making text-based detection unreliable.

The study shows that AI-generated posts can mimic human posts effectively. Current text-based detectors are not reliable for distinguishing between them. This undermines trust in disaster social sensing, which relies on accurate situational awareness.

SOURCE     arXiv cs.CY (Computers and Society) (tier 1)
   ↓
DOCUMENT   8d34f580-5257-438c-a045-4d642534bb2b
           https://arxiv.org/abs/2609.35821
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  3 runs, methodology 1.0.1
   ↓
SCORE      -58  (Adverse)
   ↓
CONFIDENCE 82%