Study Finds Humans and AI Struggle to Detect Synthetic Images in Legal Evidence
Researchers created a dataset of authentic and AI-generated legal evidence images and tested both humans and multimodal AI models on their ability to distinguish them. Humans achieved 64.8% accuracy overall but performed at chance level (around 50%) on images from the most advanced generators, while AI models never falsely flagged authentic images but missed most synthetic ones from harder generators. The findings suggest neither humans nor current AI systems can reliably authenticate visual evidence alone, raising concerns for legal proceedings.
A new study published on arXiv examined how well humans and frontier multimodal large language models (MLLMs) can detect AI-generated images masquerading as legal evidence. The researchers built SLED-1400, a dataset containing 200 authentic evidence photographs paired with 1,200 synthetic images created by six text-to-image generators across ten evidence categories. They tested 136 lay participants and four state-of-the-art MLLMs (GPT-5.1, Gemini-3-Pro, Gemini-3-Flash, and Qwen3-VL-235B) using identical stimuli and response formats. Results showed human accuracy dropped to near-chance levels (48.5-51%) when facing outputs from the strongest generators, while MLLMs achieved perfect specificity (never falsely flagging authentic images) but detected only 5.9% of synthetic outputs from the most advanced generator. The authors conclude that visual evidence in legal proceedings should be treated as inherently contestable and recommend a multi-layered approach combining trained human review, MLLM screening, and technical provenance infrastructure like C2PA Content Credentials.
What's missing
The study does not discuss potential limitations such as: whether results generalize beyond the specific evidence categories tested; how performance might differ with legal professionals versus lay participants; whether detection improves with explicit training; or how quickly detection methods can adapt as generative models advance. The paper also does not address the computational cost or practical feasibility of implementing the recommended multi-layered authentication approach in real legal settings.
What different sources said
- arXiv cs.AICenter
The CIFAR Synthetic Evidence Corpus for Detecting AI-Generated Evidence
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.