Pith. sign in

REVIEW 6 cited by

BadGPT: Exploring Security Vulnerabilities of ChatGPT via Backdoor Attacks to InstructGPT

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2304.12298 v1 pith:IKPBBXOV submitted 2023-02-21 cs.CR cs.AI

Jiawen Shi , Yixin Liu , Pan Zhou , Lichao Sun This is my paper
classification cs.CRcs.AI
keywords backdoorbadgptfine-tuninglanguagemodelchatgptinstructgptmodels
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recently, ChatGPT has gained significant attention in research due to its ability to interact with humans effectively. The core idea behind this model is reinforcement learning (RL) fine-tuning, a new paradigm that allows language models to align with human preferences, i.e., InstructGPT. In this study, we propose BadGPT, the first backdoor attack against RL fine-tuning in language models. By injecting a backdoor into the reward model, the language model can be compromised during the fine-tuning stage. Our initial experiments on movie reviews, i.e., IMDB, demonstrate that an attacker can manipulate the generated text through BadGPT.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Corruption-robust Offline Multi-agent Reinforcement Learning From Human Feedback

    cs.LG 2026-03 unverdicted novelty 7.0 of 10

    Introduces robust estimators for linear Markov games in offline MARLHF that achieve O(ε^{1-o(1)}) or O(√ε) bounds on Nash or CCE gaps under uniform or unilateral coverage.

  2. Decision-Level Hijacking: Injecting Cognitive Bias into Large Language Models via Bit-Flip Attacks

    cs.CR 2026-07 conditional novelty 6.0 of 10

    A handful of weight-bit flips (as few as 12) can bias LLM outputs toward a target entity or stance, with limited effect on non-target tasks and output distributions.

  3. GradSentry: Gradient Spectral Entropy for Backdoor Sample Filtering in Large Language Model Fine-Tuning

    cs.CR 2026-05 unverdicted novelty 6.0 of 10

    GradSentry filters backdoor samples via spectral entropy of gradients, effective at 1-90% poison ratios without clustering for both LoRA and full fine-tuning.

  4. Stealthy Backdoor Attacks against LLMs Based on Natural Style Triggers

    cs.CR 2026-04 unverdicted novelty 6.0 of 10

    BadStyle creates stealthy backdoors in LLMs by poisoning samples with imperceptible style triggers and using an auxiliary loss to stabilize payload injection, achieving high attack success rates across multiple models...

  5. Industry Practitioners Perspectives on AI Model Quality: Perceptions, Challenges, and Solutions

    cs.SE 2024-02 unverdicted novelty 4.0 of 10

    Industry AI practitioners view model quality through nine attributes with context-dependent priorities, where data imbalance is a key challenge addressed by strategies like active learning, as confirmed by interviews ...

  6. Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey

    cs.CR 2024-09 unverdicted novelty 2.0 of 10

    Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.

Pith tools