Pith. sign in

Ignore previous prompt: Attack techniques for language models (2022)

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

method 1

citation-polarity summary

fields

cs.CR 1

years

2025 1

verdicts

CONDITIONAL 1

roles

method 1

polarities

use method 1

representative citing papers

Privacy and Security Threat for OpenAI GPTs

cs.CR · 2025-06-04 · conditional · novelty 6.0

A large-scale study finds that over 98.8% of sampled OpenAI custom GPTs leak their system instructions to crafted adversarial prompts, and hundreds of GPTs transmit user conversation data to third parties.

citing papers explorer

Showing 1 of 1 citing paper.

  • Privacy and Security Threat for OpenAI GPTs cs.CR · 2025-06-04 · conditional · none · ref 4

    A large-scale study finds that over 98.8% of sampled OpenAI custom GPTs leak their system instructions to crafted adversarial prompts, and hundreds of GPTs transmit user conversation data to third parties.