Library

GPT in Sheep's Clothing: The Risk of Customized GPTs

In November 2023, OpenAI introduced a new service allowing users to create custom versions of ChatGPT (GPTs) by using specific instructions and knowledge to guide the model's behavior. We aim to raise awareness of the fact that GPTs can be used maliciously, posing privacy and security risks to their users.

cs.CRcs.AI
Source

Paper

Similar papers

Assessing Prompt Injection Risks in 200+ Custom GPTsJiahao Yu, Yuhang Wu, Dong Shu et al.2023 · arXiv (Cornell University) In Library
ChatGPT - Information Security OverviewGabriela TOD-RĂILEANU, Sabina-Daniela Axinte2023 In Library
Privacy Perceptions of Custom GPTs by Users and CreatorsRongjun Ma, Caterina Maidhof, Juan Carlos Carrillo et al.2025 In Library
ChatGPT: friend or foe?The Lancet Digital Health2023 · The Lancet Digital Health In Library
Chat GPT – Developments, Features, Impacts, and Ethical ChallengesTimothy Raboen Kwalanta2026 · Zenodo (CERN European Organization for Nuclear Research) In Library
Chat GPT – Developments, Features, Impacts, and Ethical ChallengesTimothy Raboen Kwalanta2026 · Zenodo (CERN European Organization for Nuclear Research) In Library
Chat GPT – Developments, Features, Impacts, and Ethical ChallengesTimothy Raboen Kwalanta2026 · Zenodo (CERN European Organization for Nuclear Research) In Library
ChatGPT OpenAI Unleashed: Ethical and Effective Strategies for ResearchersT Manjula, Indra Kumar2024 · Journal of Library and Information Communication Technology In Library
Security Concerns With ChatGPT and Other AI ToolsA. Anju, R. Adline Freeda, Krithikaa Venket2024 · Advances in computational intelligence and robotics book series In Library
An evaluation of the impact of ChatGPT on network securityQizhong Zheng2024 · Applied and Computational Engineering In Library
© 2026 NYSGPT2525 LLC