pith. machine review for the scientific record. sign in
Pith Number

pith:2C5FQKYP

pith:2023:2C5FQKYPGW24CFMA4OGXYFHUNI
not attested not anchored not stored refs resolved

Low-Resource Languages Jailbreak GPT-4

Cristina Menghini, Stephen H. Bach, Zheng-Xin Yong

Translating harmful English prompts into low-resource languages lets GPT-4 provide actionable advice for bad goals 79 percent of the time.

arxiv:2310.02446 v2 · 2023-10-03 · cs.CL · cs.AI · cs.CR · cs.LG

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

On the AdvBenchmark, GPT-4 engages with the unsafe translated inputs and provides actionable items that can get the users towards their harmful goals 79% of the time, which is on par with or even surpassing state-of-the-art jailbreaking attacks.

C2weakest assumption

That automatic translations preserve the original unsafe intent without introducing detectable artifacts or triggering the model's cross-lingual safety mechanisms, and that results on the tested benchmark generalize to other prompts and models.

C3one line summary

Translating unsafe inputs to low-resource languages jailbreaks GPT-4 at rates on par with or exceeding state-of-the-art attacks.

References

55 extracted · 55 resolved · 13 Pith anchors

[1] Jigsaw multilingual toxic comment classification, 2020 2020
[2] Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback 2022 · arXiv:2204.05862
[3] Constitutional AI: Harmlessness from AI Feedback 2022 · arXiv:2212.08073
[4] A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity 2023 · arXiv:2302.04023
[5] X., Macherey, K., Krikun, M., Wang, P., Gutkin, A., Shah, A., Huang, Y., Chen, Z., Wu, Y., and Hughes, M 2022

Formal links

2 machine-checked theorem links

Cited by

20 papers in Pith

Receipt and verification
First computed 2026-05-17T23:38:14.486235Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

d0ba582b0f35b5c11580e38d7c14f46a19141ba58e5252d5433334f7f24ce5cc

Aliases

arxiv: 2310.02446 · arxiv_version: 2310.02446v2 · doi: 10.48550/arxiv.2310.02446 · pith_short_12: 2C5FQKYPGW24 · pith_short_16: 2C5FQKYPGW24CFMA · pith_short_8: 2C5FQKYP
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/2C5FQKYPGW24CFMA4OGXYFHUNI \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: d0ba582b0f35b5c11580e38d7c14f46a19141ba58e5252d5433334f7f24ce5cc
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "30e80a9acc7c61f31bb9748fe5e9274c34733005765d8178bff8e059c3b3b223",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CR",
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2023-10-03T21:30:56Z",
    "title_canon_sha256": "33bbf057176543a71f37ad4685c2ce1ea6fbd47b59450069528b23f272988709"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2310.02446",
    "kind": "arxiv",
    "version": 2
  }
}