Policy-as-Code architectural approach to protect LLM-inference pipelines from prompt injection attacks

Архітектурний підхід Policy-as-Code для захисту LLM-інференс пайплайнів від атак prompt injection

ITSec-2026

1
Oleksandr Vakhula
"Lviv Polytechnic" National University
Handle 17.itsec/article.2026851
Conference ITSec-2026
Article PDF p. 138
Proceedings Proceedings of ITSec-2026 (PDF) ↗

Cite

Vakhula, O. (2026). Policy-as-Code architectural approach to protect LLM-inference pipelines from prompt injection attacks. ITSec-2026. 17.itsec/article.2026851

@inproceedings{itsec2026_851,
  author    = {Vakhula, O.},
  title     = {Policy-as-Code architectural approach to protect LLM-inference pipelines from prompt injection attacks},
  booktitle = {ITSec-2026},
  year      = {2026},
  note      = {17.itsec/article.2026851}
}
Suggest an edit

Suggest an edit

Spotted a mistake or want to update some information? Fill in the form below. All submissions go through moderation.

Current value
Архітектурний підхід Policy-as-Code для захисту LLM-інференс пайплайнів від атак prompt injection
Proposed value
Current value
Policy-as-Code architectural approach to protect LLM-inference pipelines from prompt injection attacks
Proposed value
Current value
— (not set)
Proposed value
Current value
— (not set)
Proposed value
Current value
— (not set)
Proposed value
Current value
— (not set)
Proposed value
Explain the changes, or include information not covered by the form.