{
 "id": "AIR-2023-0012",
 "slug": "mobile-deep-learning-apps-vulnerable-to-neural-payload-injec",
 "title": "Mobile Deep Learning Apps Vulnerable to Neural Payload Injection",
 "summary": "Researchers found that many deep learning models in Google Play apps are vulnerable to backdoor attacks via neural payload injection. The empirical study identified multiple critical applications affected by this vulnerability.",
 "type": "attack_on_ai",
 "categories": [
  "data_poisoning",
  "supply_chain"
 ],
 "severity": 1,
 "targets": [
  "open_weight_model",
  "enterprise_deployment"
 ],
 "products": [
  "Mobile Deep Learning Apps"
 ],
 "incident_date": "2023-03-31",
 "incident_date_precision": "day",
 "first_reported": "2023-03-31",
 "sources": [
  {
   "url": "https://github.com/avidml/avid-db/blob/main/vulnerabilities/2023/AVID-2023-V013.json",
   "publisher": "AVID",
   "title": "AVID-2023-V013: Deep learning models are increasingly used in mobile applications as critical components.",
   "published_at": "2023-03-31"
  },
  {
   "url": "https://atlas.mitre.org/studies/AML.CS0013",
   "publisher": "MITRE ATLAS",
   "title": "Backdoor Attack on Deep Learning Models in Mobile Apps",
   "published_at": "2023-03-31"
  },
  {
   "url": "https://arxiv.org/abs/2101.06896",
   "publisher": "arXiv",
   "title": "DeepPayload: Black-box Backdoor Attack on Deep Learning Models through Neural Payload Injection",
   "published_at": "2023-03-31"
  }
 ],
 "source_count": 3,
 "confidence": 0.9,
 "status": "published",
 "origin": "seed_avid",
 "tags": [
  "backdoor",
  "data_poisoning",
  "mobile_apps"
 ],
 "created_at": "2026-10-18T06:56:23.100Z",
 "updated_at": "2026-10-18T06:56:23.100Z"
}