ML Security Papers
Browse Stats

Research landscape

3669 papers across 20 OWASP categories

3669
papers
1139
attack (31%)
1734
defense (47%)
28
this month

Monthly publications

OWASP ML Top 10

ML01 Input Manipulation Attack
839
ML02 Data Poisoning Attack
291
ML03 Model Inversion Attack
244
ML04 Membership Inference Attack
211
ML05 Model Theft
165
ML06 AI Supply Chain Attacks
47
ML07 Transfer Learning Attack
86
ML08 Model Skewing
7
ML09 Output Integrity Attack
742
ML10 Model Poisoning
337

OWASP LLM Top 10

LLM01 Prompt Injection
1223
LLM02 Insecure Output Handling
0
LLM03 Training Data Poisoning
43
LLM04 Model Denial of Service
23
LLM05 Supply Chain Vulnerabilities
0
LLM06 Sensitive Information Disclosure
211
LLM07 Insecure Plugin Design
111
LLM08 Excessive Agency
219
LLM09 Overreliance
0
LLM10 Model Theft
53

Domains

nlp 2027
vision 1484
multimodal 547
generative 444
federated-learning 268
audio 176
graph 93
tabular 91
reinforcement-learning 78
timeseries 38

Paper types