Home / Current Issue / Paper 1725422
Prompt Injection and Jailbreak Attacks on LLM Applications: A Review of Attacks, Defenses and Risk-Adaptive Protection
Subject area: Science,Engineering and Technology · Area of research: LLM Applications
Abstract
Large language models (LLMs) are now embedded in chatbots, coding assistants, search tools and agents, and this has exposed a new class of security problems in which attackers use ordinary text to override a developer’s instructions or to push a model past its safety training. This paper reviews the literature on these problems, focusing on prompt injection and jailbreak attacks. It organises attacks into a taxonomy (direct and indirect injection, manual, optimisation-based, LLM-driven, obfuscated, multi-turn, multimodal and training-time attacks), organises defenses into eight layers from alignment-time training to system-level controls, and summarises the benchmarks and metrics used to compare them. It then examines a theme that is discussed less often than attack and defense strength, namely the cost of defending: most published defenses are applied uniformly to every request, while research on LLM routing and adaptive moderation shows that matching effort to difficulty can save cost. The paper discusses risk-proportional defense as a design direction, and closes with open problems, including adaptive attackers, evaluation validity, over-refusal, multi-model deployments and agents. The review draws on more than one hundred papers and resources published between 2020 and 2026.
Keywords
adaptive defense, jailbreak, large language models, LLM security, prompt injection
References
[1] Brown TB, et al. Language models are few-shot learners. In: Advances in Neural Information Processing Systems (NeurIPS); 2020. arXiv:2005.14165.
[2] Ouyang L, et al. Training language models to follow instructions with human feedback. In: Advances in Neural Information Processing Systems (NeurIPS); 2022. arXiv:2203.02155.
[3] Bai Y, et al. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv [Preprint]. 2022. arXiv:2204.05862.
[4] OpenAI. GPT-4 technical report. arXiv [Preprint]. 2023. arXiv:2303.08774.
[5] Touvron H, et al. LLaMA: open and efficient foundation language models. arXiv [Preprint]. 2023. arXiv:2302.13971.
[6] Touvron H, et al. Llama 2: open foundation and fine-tuned chat models. arXiv [Preprint]. 2023. arXiv:2307.09288.
[7] Jiang AQ, et al. Mistral 7B. arXiv [Preprint]. 2023. arXiv:2310.06825.
[8] Jiang AQ, et al. Mixtral of experts. arXiv [Preprint]. 2024. arXiv:2401.04088.
[9] Gemma Team. Gemma: open models based on Gemini research and technology. arXiv [Preprint]. 2024. arXiv:2403.08295.
[10] Gemini Team. Gemini: a family of highly capable multimodal models. arXiv [Preprint]. 2023. arXiv:2312.11805.
[11] DeepSeek-AI. DeepSeek-V3 technical report. arXiv [Preprint]. 2024. arXiv:2412.19437.
[12] DeepSeek-AI. DeepSeek-R1: incentivizing reasoning capability in LLMs via reinforcement learning. arXiv [Preprint]. 2025. arXiv:2501.12948.
[13] Dubey A, et al. The Llama 3 herd of models. arXiv [Preprint]. 2024. arXiv:2407.21783.
[14] Qwen Team. Qwen2.5 technical report. arXiv [Preprint]. 2024. arXiv:2412.15115.
[15] Lewis P, et al. Retrieval-augmented generation for knowledge-intensive NLP tasks. In: Advances in Neural Information Processing Systems (NeurIPS); 2020. arXiv:2005.11401.
[16] Perez F, Ribeiro I. Ignore previous prompt: attack techniques for language models. In: NeurIPS 2022 ML Safety Workshop; 2022. arXiv:2211.09527.
[17] Liu Y, Jia Y, Geng R, Jia J, Gong NZ. Formalizing and benchmarking prompt injection attacks and defenses. In: USENIX Security; 2024. arXiv:2310.12815.
[18] Wei A, Haghtalab N, Steinhardt J. Jailbroken: how does LLM safety training fail? In: Advances in Neural Information Processing Systems (NeurIPS); 2023. arXiv:2307.02483.
[19] Zou A, et al. Universal and transferable adversarial attacks on aligned language models. arXiv [Preprint]. 2023. arXiv:2307.15043.
[20] OWASP GenAI Security Project. OWASP top 10 for large language model applications [Internet]. Available from: https://owasp.org
[21] Vassilev A, Oprea A, Fordyce A, Anderson H. Adversarial machine learning: a taxonomy and terminology of attacks and mitigations. NIST AI 100-2e2023. Gaithersburg, MD: National Institute of Standards and Technology; 2024. https://doi.org/10.6028/NIST.AI.100-2e2023
[22] MITRE. MITRE ATLAS: adversarial threat landscape for artificial-intelligence systems [Internet]. Available from: https://atlas.mitre.org
[23] Yao Y, et al. A survey on large language model (LLM) security and privacy: the good, the bad, and the ugly. High-Confidence Computing. 2024. arXiv:2312.02003.
[24] Das BC, Amini MH, Wu Y. Security and privacy challenges of large language models: a survey. ACM Computing Surveys. 2025. arXiv:2402.00888.
[25] Shayegani E, et al. Survey of vulnerabilities in large language models revealed by adversarial attacks. arXiv [Preprint]. 2023. arXiv:2310.10844.
[26] Schwinn L, et al. Adversarial attacks and defenses in large language models: old and new threats. arXiv [Preprint]. 2023. arXiv:2310.19737.
[27] Yi S, et al. Jailbreak attacks and defenses against large language models: a survey. arXiv [Preprint]. 2024. arXiv:2407.04295.
[28] Xu Z, et al. LLM jailbreak attack versus defense techniques: a comprehensive study. arXiv [Preprint]. 2024. arXiv:2402.13457.
[29] Liu Y, et al. Trustworthy LLMs: a survey and guideline for evaluating large language models’ alignment. arXiv [Preprint]. 2023. arXiv:2308.05374.
[30] Zhan Q, Liang Z, Ying Z, Kang D. InjecAgent: benchmarking indirect prompt injections in tool-integrated large language model agents. In: Findings of ACL; 2024. arXiv:2403.02691.
[31] Debenedetti E, et al. AgentDojo: a dynamic environment to evaluate prompt injection attacks and defenses for LLM agents. In: NeurIPS 2024 Datasets and Benchmarks; 2024. arXiv:2406.13352.
[32] Ollama. Run large language models locally [Internet]. Available from: https://ollama.com
[33] Hugging Face. Hugging Face Hub: models and datasets [Internet]. Available from: https://huggingface.co
[34] Wolf T, et al. Transformers: state-of-the-art natural language processing. In: EMNLP 2020 System Demonstrations; 2020. arXiv:1910.03771.
[35] Bai Y, et al. Constitutional AI: harmlessness from AI feedback. arXiv [Preprint]. 2022. arXiv:2212.08073.
[36] Rafailov R, et al. Direct preference optimization: your language model is secretly a reward model. In: Advances in Neural Information Processing Systems (NeurIPS); 2023. arXiv:2305.18290.
[37] Ganguli D, et al. Red teaming language models to reduce harms: methods, scaling behaviors, and lessons learned. arXiv [Preprint]. 2022. arXiv:2209.07858.
[38] Perez E, et al. Red teaming language models with language models. In: EMNLP; 2022. arXiv:2202.03286.
[39] Perez E, et al. Discovering language model behaviors with model-written evaluations. In: Findings of ACL; 2023. arXiv:2212.09251.
[40] Bommasani R, et al. On the opportunities and risks of foundation models. arXiv [Preprint]. 2021. arXiv:2108.07258.
[41] Weidinger L, et al. Ethical and social risks of harm from language models. arXiv [Preprint]. 2021. arXiv:2112.04359.
[42] Hendrycks D, et al. Unsolved problems in ML safety. arXiv [Preprint]. 2021. arXiv:2109.13916.
[43] Cui T, et al. Risk taxonomy, mitigation, and assessment benchmarks of large language model systems. arXiv [Preprint]. 2024. arXiv:2401.05778.
[44] Hui B, et al. PLeak: prompt leaking attacks against large language model applications. In: ACM CCS; 2024. arXiv:2405.06823.
[45] Zhang Y, Ippolito D. Effective prompt extraction from language models. arXiv [Preprint]. 2023. arXiv:2307.06865.
[46] Zhu S, et al. AutoDAN: interpretable gradient-based adversarial attacks on large language models. In: COLM; 2024. arXiv:2310.15140.
[47] Andriushchenko M, Croce F, Flammarion N. Jailbreaking leading safety-aligned LLMs with simple adaptive attacks. arXiv [Preprint]. 2024. arXiv:2404.02151.
[48] Carlini N, et al. Poisoning web-scale training datasets is practical. In: IEEE S&P; 2024. arXiv:2302.10149.
[49] Wan A, et al. Poisoning language models during instruction tuning. In: International Conference on Machine Learning (ICML); 2023. arXiv:2305.00944.
[50] Zou W, et al. PoisonedRAG: knowledge corruption attacks to retrieval-augmented generation of large language models. In: USENIX Security; 2025. arXiv:2402.07867.
[51] Greshake K, et al. Not what you’ve signed up for: compromising real-world LLM-integrated applications with indirect prompt injection. In: AISec; 2023. arXiv:2302.12173.
[52] Rossi S, et al. An early categorization of prompt injection attacks on large language models. arXiv [Preprint]. 2024. arXiv:2402.00898.
[53] Yi J, et al. Benchmarking and defending against indirect prompt injection attacks on large language models. arXiv [Preprint]. 2023. arXiv:2312.14197.
[54] Liu Y, et al. Jailbreaking ChatGPT via prompt engineering: an empirical study. arXiv [Preprint]. 2023. arXiv:2305.13860.
[55] Shen X, Chen Z, Backes M, Shen Y, Zhang Y. Do anything now: characterizing and evaluating in-the-wild jailbreak prompts on large language models. In: ACM CCS; 2024. arXiv:2308.03825.
[56] Mazeika M, et al. HarmBench: a standardized evaluation framework for automated red teaming and robust refusal. In: International Conference on Machine Learning (ICML); 2024. arXiv:2402.04249.
[57] Branch HJ, et al. Evaluating the susceptibility of pre-trained language models via handcrafted adversarial examples. arXiv [Preprint]. 2022. arXiv:2209.02128.
[58] Liu Y, et al. Prompt injection attack against LLM-integrated applications. arXiv [Preprint]. 2023. arXiv:2306.05499.
[59] Liu X, et al. Automatic and universal prompt injection attacks against large language models. arXiv [Preprint]. 2024. arXiv:2403.04957.
[60] Pedro R, et al. From prompt injections to SQL injection attacks: how protected is your LLM-integrated web application? arXiv [Preprint]. 2023. arXiv:2308.01990.
[61] Shi J, et al. Optimization-based prompt injection attack to LLM-as-a-judge. In: ACM CCS; 2024. arXiv:2403.17710.
[62] Toyer S, et al. Tensor Trust: interpretable prompt injection attacks from an online game. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2311.01011.
[63] Schulhoff S, et al. Ignore this title and HackAPrompt: exposing systemic vulnerabilities of LLMs through a global scale prompt hacking competition. In: EMNLP; 2023. arXiv:2311.16119.
[64] Ruan Y, et al. Identifying the risks of LM agents with an LM-emulated sandbox. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2309.15817.
[65] Zhuo TY, et al. Red teaming ChatGPT via jailbreaking: bias, robustness, reliability and toxicity. arXiv [Preprint]. 2023. arXiv:2301.12867.
[66] Kang D, et al. Exploiting programmatic behavior of LLMs: dual-use through standard security attacks. arXiv [Preprint]. 2023. arXiv:2302.05733.
[67] Wolf Y, et al. Fundamental limitations of alignment in large language models. In: International Conference on Machine Learning (ICML); 2024. arXiv:2304.11082.
[68] Carlini N, et al. Are aligned neural networks adversarially aligned? In: Advances in Neural Information Processing Systems (NeurIPS); 2023. arXiv:2306.15447.
[69] Liu X, Xu N, Chen M, Xiao C. AutoDAN: generating stealthy jailbreak prompts on aligned large language models. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2310.04451.
[70] Lapid R, Langberg R, Sipper M. Open Sesame! Universal black box jailbreaking of large language models. arXiv [Preprint]. 2023. arXiv:2309.01446.
[71] Chao P, et al. Jailbreaking black box large language models in twenty queries. arXiv [Preprint]. 2023. arXiv:2310.08419.
[72] Mehrotra A, et al. Tree of attacks: jailbreaking black-box LLMs automatically. In: Advances in Neural Information Processing Systems (NeurIPS); 2024. arXiv:2312.02119.
[73] Yu J, Lin X, Yu Z, Xing X. GPTFuzzer: red teaming large language models with auto-generated jailbreak prompts. arXiv [Preprint]. 2023. arXiv:2309.10253.
[74] Deng G, et al. MasterKey: automated jailbreak across multiple large language model chatbots. In: NDSS; 2024. arXiv:2307.08715.
[75] Yuan Y, et al. GPT-4 is too smart to be safe: stealthy chat with LLMs via cipher. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2308.06463.
[76] Yong ZX, Menghini C, Bach SH. Low-resource languages jailbreak GPT-4. arXiv [Preprint]. 2023. arXiv:2310.02446.
[77] Deng Y, et al. Multilingual jailbreak challenges in large language models. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2310.06474.
[78] Li X, et al. DeepInception: hypnotize large language model to be jailbreaker. arXiv [Preprint]. 2023. arXiv:2311.03191.
[79] Huang Y, et al. Catastrophic jailbreak of open-source LLMs via exploiting generation. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2310.06987.
[80] Russinovich M, Salem A, Eldan R. Great, now write an article about that: the Crescendo multi-turn LLM jailbreak attack. In: USENIX Security; 2025. arXiv:2404.01833.
[81] Zeng Y, et al. How Johnny can persuade LLMs to jailbreak them: rethinking persuasion to challenge AI safety by humanizing LLMs. In: ACL; 2024. arXiv:2401.06373.
[82] Wei Z, Wang Y, Wang Y. Jailbreak and guard aligned language models with only few in-context demonstrations. arXiv [Preprint]. 2023. arXiv:2310.06387.
[83] Qi X, et al. Visual adversarial examples jailbreak aligned large language models. In: AAAI; 2024. arXiv:2306.13213.
[84] Bagdasaryan E, et al. Abusing images and sounds for indirect instruction injection in multi-modal LLMs. arXiv [Preprint]. 2023. arXiv:2307.10490.
[85] Shu M, et al. On the exploitability of instruction tuning. In: Advances in Neural Information Processing Systems (NeurIPS); 2023. arXiv:2306.17194.
[86] Qi X, et al. Fine-tuning aligned language models compromises safety, even when users do not intend to! In: International Conference on Learning Representations (ICLR); 2024. arXiv:2310.03693.
[87] Wallace E, et al. The instruction hierarchy: training LLMs to prioritize privileged instructions. arXiv [Preprint]. 2024. arXiv:2404.13208.
[88] Chen S, Piet J, Sitawarin C, Wagner D. StruQ: defending against prompt injection with structured queries. In: USENIX Security; 2025. arXiv:2402.06363.
[89] Piet J, et al. Jatmo: prompt injection defense by task-specific finetuning. arXiv [Preprint]. 2023. arXiv:2312.17673.
[90] Zou A, et al. Improving alignment and robustness with circuit breakers. In: Advances in Neural Information Processing Systems (NeurIPS); 2024. arXiv:2406.04313.
[91] Zou A, et al. Representation engineering: a top-down approach to AI transparency. arXiv [Preprint]. 2023. arXiv:2310.01405.
[92] Xie Y, et al. Defending ChatGPT against jailbreak attack via self-reminders. Nature Machine Intelligence. 2023;5. https://doi.org/10.1038/s42256-023-00765-8.
[93] Hines K, et al. Defending against indirect prompt injection attacks with spotlighting. arXiv [Preprint]. 2024. arXiv:2403.14720.
[94] Jain N, et al. Baseline defenses for adversarial attacks against aligned language models. arXiv [Preprint]. 2023. arXiv:2309.00614.
[95] Alon G, Kamfonas M. Detecting language model attacks with perplexity. arXiv [Preprint]. 2023. arXiv:2308.14132.
[96] He P, et al. DeBERTa: decoding-enhanced BERT with disentangled attention. arXiv [Preprint]. 2020. arXiv:2006.03654.
[97] Hu X, Chen PY, Ho TY. Gradient Cuff: detecting jailbreak attacks on large language models by exploring refusal loss landscapes. In: Advances in Neural Information Processing Systems (NeurIPS); 2024. arXiv:2403.00867.
[98] Robey A, Wong E, Hassani H, Pappas GJ. SmoothLLM: defending large language models against jailbreaking attacks. arXiv [Preprint]. 2023. arXiv:2310.03684.
[99] Ji J, et al. Defending large language models against jailbreak attacks via semantic smoothing. arXiv [Preprint]. 2024. arXiv:2402.16192.
[100] Kumar A, et al. Certifying LLM safety against adversarial prompting. arXiv [Preprint]. 2023. arXiv:2309.02705.
[101] Cao B, et al. Defending against alignment-breaking attacks via robustly aligned LLM. In: ACL; 2024. arXiv:2309.14348.
[102] Zhou A, Li B, Wang H. Robust prompt optimization for defending language models against jailbreaking attacks. In: Advances in Neural Information Processing Systems (NeurIPS); 2024. arXiv:2401.17263.
[103] Xu Z, et al. SafeDecoding: defending against jailbreak attacks via safety-aware decoding. In: ACL; 2024. arXiv:2402.08983.
[104] Li Y, et al. RAIN: your language models can align themselves without finetuning. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2309.07124.
[105] Phute M, et al. LLM self defense: by self examination, LLMs know they are being tricked. arXiv [Preprint]. 2023. arXiv:2308.07308.
[106] Zhang Y, et al. Intention analysis makes LLMs a good jailbreak defender. arXiv [Preprint]. 2024. arXiv:2401.06561.
[107] Zhang Z, et al. Defending large language models against jailbreaking attacks through goal prioritization. In: ACL; 2024. arXiv:2311.09096.
[108] Wang Z, et al. Self-Guard: empower the LLM to safeguard itself. In: NAACL; 2024. arXiv:2310.15851.
[109] Markov T, et al. A holistic approach to undesired content detection in the real world. In: AAAI; 2023. arXiv:2208.03274.
[110] Gehman S, et al. RealToxicityPrompts: evaluating neural toxic degeneration in language models. In: Findings of EMNLP; 2020. arXiv:2009.11462.
[111] Hartvigsen T, et al. ToxiGen: a large-scale machine-generated dataset for adversarial and implicit hate speech detection. In: ACL; 2022. arXiv:2203.09509.
[112] Lin Z, et al. ToxicChat: unveiling hidden challenges of toxicity detection in real-world user-AI conversation. In: Findings of EMNLP; 2023. arXiv:2310.17389.
[113] Inan H, et al. Llama Guard: LLM-based input-output safeguard for human-AI conversations. Meta AI; 2023. arXiv:2312.06674.
[114] Han S, et al. WildGuard: open one-stop moderation tools for safety risks, jailbreaks, and refusals of LLMs. In: Advances in Neural Information Processing Systems (NeurIPS); 2024. arXiv:2406.18495.
[115] Zeng W, et al. ShieldGemma: generative AI content moderation based on Gemma. arXiv [Preprint]. 2024. arXiv:2407.21772.
[116] Padhi I, et al. Granite Guardian. arXiv [Preprint]. 2024. arXiv:2412.07724.
[117] Sharma M, et al. Constitutional classifiers: defending against universal jailbreaks across thousands of hours of red teaming. arXiv [Preprint]. 2025. arXiv:2501.18837.
[118] Rebedea T, et al. NeMo Guardrails: a toolkit for controllable and safe LLM applications with programmable rails. In: EMNLP 2023 System Demonstrations; 2023. arXiv:2310.10501.
[119] Dong Y, et al. Building guardrails for large language models. In: International Conference on Machine Learning (ICML); 2024. arXiv:2402.01822.
[120] Bhatt M, et al. Purple Llama CyberSecEval: a secure coding benchmark for language models. arXiv [Preprint]. 2023. arXiv:2312.04724.
[121] Bhatt M, et al. CyberSecEval 2: a wide-ranging cybersecurity evaluation suite for large language models. arXiv [Preprint]. 2024. arXiv:2404.13161.
[122] Chao P, et al. JailbreakBench: an open robustness benchmark for jailbreaking large language models. In: NeurIPS 2024 Datasets and Benchmarks; 2024. arXiv:2404.01318.
[123] Souly A, et al. A StrongREJECT for empty jailbreaks. In: NeurIPS 2024 Datasets and Benchmarks; 2024. arXiv:2402.10260.
[124] Wang B, et al. DecodingTrust: a comprehensive assessment of trustworthiness in GPT models. In: Advances in Neural Information Processing Systems (NeurIPS); 2023. arXiv:2306.11698.
[125] Zhang Z, et al. SafetyBench: evaluating the safety of large language models. In: ACL; 2024. arXiv:2309.07045.
[126] Huang Y, et al. TrustLLM: trustworthiness in large language models. In: International Conference on Machine Learning (ICML); 2024. arXiv:2401.05561.
[127] Bhardwaj R, Poria S. Red-teaming large language models using chain of utterances for safety-alignment. arXiv [Preprint]. 2023. arXiv:2308.09662.
[128] Ghosh S, Varshney P, Galinkin E, Parisien C. Aegis: online adaptive AI content safety moderation with ensemble of LLM experts. arXiv [Preprint]. 2024. arXiv:2404.05993.
[129] Röttger P, et al. XSTest: a test suite for identifying exaggerated safety behaviours in large language models. In: NAACL; 2024. arXiv:2308.01263.
[130] Cui J, et al. OR-Bench: an over-refusal benchmark for large language models. arXiv [Preprint]. 2024. arXiv:2405.20947.
[131] Zheng L, et al. Judging LLM-as-a-judge with MT-Bench and Chatbot Arena. In: Advances in Neural Information Processing Systems (NeurIPS); 2023. arXiv:2306.05685.
[132] Chen L, Zaharia M, Zou J. FrugalGPT: how to use large language models while reducing cost and improving performance. arXiv [Preprint]. 2023. arXiv:2305.05176.
[133] Ong I, et al. RouteLLM: learning to route LLMs with preference data. arXiv [Preprint]. 2024. arXiv:2406.18665.
[134] Hu QJ, et al. RouterBench: a benchmark for multi-LLM routing system. arXiv [Preprint]. 2024. arXiv:2403.12031.
[135] Ding D, et al. Hybrid LLM: cost-efficient and quality-aware query routing. In: International Conference on Learning Representations (ICLR); 2024. arXiv:2404.14618.
[136] Aggarwal P, et al. AutoMix: automatically mixing language models. In: Advances in Neural Information Processing Systems (NeurIPS); 2024. arXiv:2310.12963.
[137] Schuster T, et al. Confident adaptive language modeling. In: Advances in Neural Information Processing Systems (NeurIPS); 2022. arXiv:2207.07061.
[138] [Authors to be supplied]. AEGIS-Chat: context-aware no-regret safety moderation for real-world user-AI conversations. Preprint. 2026. [Identifier to be checked: arXiv:2602.20459].
[139] Kaggle. Kaggle datasets and notebooks [Internet]. Available from: https://kaggle.com
How to cite this paper
@article{1725422,
author = {Prabhakarn, Dr. Birjendra Gupta},
title = {Prompt Injection and Jailbreak Attacks on LLM Applications: A Review of Attacks, Defenses and Risk-Adaptive Protection},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {5},
pages = {4087-4105},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1725422.pdf},
abstract = {Large language models (LLMs) are now embedded in chatbots, coding assistants, search tools and agents, and this has exposed a new class of security problems in which attackers use ordinary text to override a developer’s instructions or to push a model past its safety training. This paper reviews the literature on these problems, focusing on prompt injection and jailbreak attacks. It organises attacks into a taxonomy (direct and indirect injection, manual, optimisation-based, LLM-driven, obfuscated, multi-turn, multimodal and training-time attacks), organises defenses into eight layers from alignment-time training to system-level controls, and summarises the benchmarks and metrics used to compare them. It then examines a theme that is discussed less often than attack and defense strength, namely the cost of defending: most published defenses are applied uniformly to every request, while research on LLM routing and adaptive moderation shows that matching effort to difficulty can save cost. The paper discusses risk-proportional defense as a design direction, and closes with open problems, including adaptive attackers, evaluation validity, over-refusal, multi-model deployments and agents. The review draws on more than one hundred papers and resources published between 2020 and 2026.},
keywords = {adaptive defense, jailbreak, large language models, LLM security, prompt injection},
month = {November},
}