Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 106 papers

Filter2

Attack type

Jailbreaks6 Prompt injection45 Poisoning and backdoors12 Detection evasion4

Attack technique

Dormant triggers4

Model risks

Reward hacking8 Dangerous agent actions25 Hallucination and fabrication4

Defense type

Detection and filtering4 Instruction–data separation23 Permissions and approval60 Monitoring and auditing4

Impact

Unauthorized actions58 Data and privacy leaks12 Output manipulation15 Availability and reliability3 Oversight failure7

System component

MCP and skills83 Memory80 RAG51 Guardrails and judges172 Monitors92 Reasoning chains77 Permissions and sandboxing93 Fine-tuning and open weights65 Vision140 Audio38 Video13 Image generation49 Embeddings13 Multimodal227

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 240papers have no tags yet and are not included in filtered results.

System component：Permissions and sandboxing System component：Video Clear all

Another 240papers do not have category tags yet and are not included in filtered results.

DefensesarXiv 2610.13733Oct 9
KVeil: First-Place Solution for the ECCV 2026 LifeGenIP Unlearnable Video Challenge
Propose KVeil, which adds perturbations to face videos to prevent unauthorized customization from reproducing identity
arXiv
2610.13733
Published
Oct 9
Layer
Prompt layer
Setting
Models and APIs
Attacker
White-box, Black-box
Evaluated models
Wan2.2-TI2V-5B, HunyuanVideo-1.5, Kling 3.0 and others4
- AttacksMalicious use
- ComponentImage generation
AbstractKVeil adapts LETHE's reference-reading disruption to the challenge's two stages, ranking first on the leaderboard in both Final columns.
Paper analysisRead analysis
DefensesarXiv 2610.12463Oct 9
Paper reviews OpenAI, Anthropic, and Google agent safety incidents and proposes the PASAC framework
Proposes PASAC and a five-layer boundary guarantee stack, turning agent evaluation overstepping into verifiable whole-system guarantees.
arXiv
2610.12463
Published
Oct 9
Layer
Application layer
Setting
AI Agent
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Paper analysisRead analysis
DefensesarXiv 2610.11754Oct 8
仅凭输出审查无法验证什么：面向研究智能体的研究契约
提出研究合同，把已批准选择绑定到执行证据并由检查器核验
arXiv
2610.11754
Published
Oct 8
Layer
Application layer
Setting
science agent
Evaluated models
OpenAI gpt-5.6-sol, OpenAI gpt-5.6-terra, OpenAI gpt-5.6-luna and others18
- DefensesPermissions and approval
- RiskReward hacking
- ComponentPermissions and sandboxing
Abstract研究合同暴露四类输出单独建不立的关系。
研究合同用来核验计算科学研究智能体是否仍在已批准研究之内，不判断科学是否为真。。
Paper analysisRead analysis
AttacksarXiv 2610.08331Oct 6
STCA：研究者提出针对自动驾驶视频 VLM 的时空一致性对抗攻击
提出 STCA，对驾驶视频 VLM 构造可迁移的时空对抗扰动
arXiv
2610.08331
Published
Oct 6
Layer
Prompt layer
Setting
Models and APIs
Attacker
No-box (offline transfer)
Evaluated models
Video-LLaVA-7B, Qwen2.5-VL-7B, Dolphins
- AttacksDetection evasion
- TechniqueGradient and representation optimization
- ComponentVideo
Paper analysisRead analysis
DefensesarXiv 2610.08082Oct 6
POLAR：面向小型工具调用智能体的可逆性护栏框架
提出 POLAR 本体门控，在执行前否决不可逆工具调用
arXiv
2610.08082
Published
Oct 6
Layer
Application layer
Setting
AI Agent
Evaluated models
qwen/qwen3-8b, meta-llama/llama-3.1-8b-instruct, nvidia/nemotron-nano-9b-v2 and others6
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract训练无关的事前可逆性门控，弱 airline 模型有收益，强模型与 retail 常回落。
POLAR 是训练无关的事前门控，用类型化本体在工具调用执行前检查能否组成候选逆序列。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.07939Oct 6
CCDF dataset: Grok Imagine, Veo 3.1, Sora 2 forge surveillance footage, ten deepfake detectors basically cannot recognize it
提出监控深伪基准 CCDF，在统一编码后评测检测器对伪造犯罪事故录像的识别
arXiv
2610.07939
Published
Oct 6
Layer
Prompt layer
Setting
Models and APIs
Evaluated models
C2P-CLIP, ForgeLens, D3-Im and others10
- DefensesDetection and filtering
- AttacksDetection evasion
- ComponentGuardrails and judges
Abstract提出消除格式捷径的监控深度伪造基准 CCDF，揭示现有检测器难以识别商用生成视频。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.07639Oct 6
HarnessSecurity-Bench horizontally evaluates the security mechanisms of six coding agent frameworks.
评测编码智能体 harness 安全机制在对抗任务中的防护效果
arXiv
2610.07639
Published
Oct 6
Layer
Application layer
Setting
coding agent
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Paper analysisRead analysis
OtherarXiv 2610.07459Oct 6
论文提出 AI Agent 声明的可审计性框架
提出 AI Agent 主张的可审计性框架与裁决规则
arXiv
2610.07459
Published
Oct 6
Layer
Application layer
Setting
AI Agent
- ComponentPermissions and sandboxing
Abstract主张须事先固定四要素。
作者给出把关于已部署智能体的主张写成可核查陈述的条件，而不是检测器或攻击方法。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.07359Oct 6
论文实测 Agent 工具调用门控栈的失败相关性
提出 nmult 并评测 Agent 动作门禁堆叠的错误相关性
arXiv
2610.07359
Published
Oct 6
Layer
Application layer
Setting
AI Agent
Evaluated models
AgentTrust runtime (174 rules), GPT-5.5 (gpt55), claude-haiku-4-5-20251001 and others7
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract论文实测发现大模型裁判间错误显著相关，作者主张按机制选型并在整栈层面度量防护增益。
论文是一项针对智能体动作拦截防御堆叠故障关联性与实际组合价值的实证测量研究。
Paper analysisRead analysis
DefensesarXiv 2610.07258Oct 6
AMU：用派生血缘门控实现企业 AI Agent 列级访问控制
提出 AMU 按推导图门控企业 Agent 共享记忆的列级访问
arXiv
2610.07258
Published
Oct 6
Layer
Application layer
Setting
AI Agent
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentMemory
Abstract在完整血缘下，AMU 按列权限放行缓存，并标出同名 KPI 冲突。
AMU 用列级推导门控共享记忆，拦截经缓存流出的无权敏感列，并标出同名 KPI 的定义冲突。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.06339Oct 5
BabelFake：面向多语言音视频 DeepFake 检测的基准数据集
构建多语言音视频 DeepFake 基准 BabelFake 并评测检测器
arXiv
2610.06339
Published
Oct 5
Layer
Model layer
Setting
Models and APIs
Evaluated models
SpeechForensics, AVH-Align, (B)RAVEn
- DefensesDetection and filtering
- AttacksDetection evasion
- ComponentAudio
AbstractBabelFake 为五语同意采集基准。
BabelFake 是经同意与 IRB 批准采集的多语音视频 DeepFake 检测基准，用来把视觉操纵和语音合成拆开比较。
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.06171Oct 5
ControlPed 生成逼真行人危险动作场景，七款端到端驾驶模型 HDScore 从 88.8 降至 47.4
提出 ControlPed 与 HazardPed，生成可控行人危险场景并评测端到端驾驶
arXiv
2610.06171
Published
Oct 5
Layer
Model layer
Setting
Embodied systems and robotics
Evaluated models
DiffusionDrive, SSR, VAD and others7
- ComponentVideo
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2610.05840Oct 5
研究：任务范围授权可将工具调用有害执行降至零
用配对回放评测任务范围授权能否阻断越权工具执行
arXiv
2610.05840
Published
Oct 5
Layer
Application layer
Setting
AI Agent
Evaluated models
Llama 3.2 3B Instruct Q4 K M, Qwen3 4B Instruct, Gemma 3 4B and others8
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Paper analysisRead analysis
DefensesarXiv 2610.05246Oct 4
研究测量策略网关跨平台命令裁决的能力边界
提出策略门裁决已实现动作，并测量跨平台命令拦截边界
arXiv
2610.05246
Published
Oct 4
Layer
Application layer
Setting
coding agent
Evaluated models
claude-opus-5, deepseek-v4.1-flash, gpt-6-astra
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Paper analysisRead analysis
DefensesarXiv 2610.06966Oct 4
APEX: proactively defending against indirect prompt injection at the LLM agent execution boundary.
提出 APEX，在执行边界用授权契约防御 Agent 间接提示注入
arXiv
2610.06966
Published
Oct 4
Layer
Application layer
Setting
AI Agent
Evaluated models
DeepSeek-V4-Flash, GLM-5.2, GPT-5.6-sol
- DefensesInstruction–data separation
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Paper analysisRead analysis
DefensesarXiv 2610.04975Oct 4
论文提出长程智能体自生子目标的运行时授权机制
提出长程工具智能体自生子目标的运行时授权机制
arXiv
2610.04975
Published
Oct 4
Layer
Application layer
Setting
AI Agent
- DefensesPermissions and approval
- RiskDangerous agent actions
- ComponentPermissions and sandboxing
Abstract用版本化目标图授权保住重规划，并挡住自生成子目标变成新权威。
长时程工具使用把目标演化变成授权状态的一部分。本文在单 principal、单根、有限结构化域中，把自生成子目标的创建、替换、同根委托和同根 join 做成可检查转移，并在提交边界重查受保护效应。
Paper analysisRead analysis
DefensesarXiv 2610.04504Oct 3
论文提出 VAL 框架：相同 ASR 的 LLM Agent 防御可能对应不同安全保证
提出 VAL 框架，并用确认门与参数沙箱约束智能体高风险工具调用
arXiv
2610.04504
Published
Oct 3
Layer
Application layer
Setting
AI Agent
Evaluated models
DeepSeek-chat, Meta Llama 3.1 8B, Kimi
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Paper analysisRead analysis
DefensesarXiv 2610.04699Oct 3
CoVer：用干预检验为 Agent 构造可判定规则边界
提出 COVER，用干预门判定 Agent 规则谓词能否自动执行
arXiv
2610.04699
Published
Oct 3
Layer
Application layer
Setting
AI Agent
Evaluated models
Claude Sonnet, Claude Opus, GPT-4o and others5
- DefensesPermissions and approval
- ComponentPermissions and sandboxing
Abstract自可判定不能向模型索取。
这篇工作测的是智能体验证器在检查之前的分流：外部规则里的哪条谓词能用固定过程解决，哪条必须交给评审。
Paper analysisRead analysis
DefensesarXiv 2610.02999Oct 2
OmniConfess：用 Token 级「忏悔」缓解全模态幻觉
提出 OmniConfess，冻结候选并按通道重打分以缓解全模态幻觉
arXiv
2610.02999
Published
Oct 2
Layer
Model layer
Setting
Models and APIs
Evaluated models
Qwen2.5-Omni-7B, Qwen3-Omni-30B-A3B, Nemotron-3-Nano-Omni-30B
- DefensesInference-time intervention
- RiskHallucination and fabrication
- ComponentVision
Paper analysisRead analysis
DefensesarXiv 2610.02861Oct 2
论文提出在 Kubernetes 上保护 AI Agent 的七层纵深防御框架与参考架构
提出 Kubernetes 运维智能体的七层纵深防御框架与参考架构
arXiv
2610.02861
Published
Oct 2
Layer
Application layer
Setting
AI Agent
- DefensesPermissions and approval
- AttacksPrompt injection
- ComponentPermissions and sandboxing
Abstract把运维智能体当不可信租户，用七层控制在完全攻破假设下约束行动。
把 Kubernetes 上的运维智能体当作不可信租户，在假定提示注入已完全控制其工具选择的前提下，用基础设施重建数据面与控制面的边界。要解决的是：智能体读日志、注解、工单或 MCP 工具描述时，这些内容可以变成它带着自身凭据执行的命令。模型对齐和分类器只降低概率。
Paper analysisRead analysis

AI Safety Daily

Read the latest issueOne issue per day, updated as events develop

All papers

Attack type

Attack technique

Model risks

Defense type

Impact

System type

Attack surface / entry point

Layer

System component

KVeil: First-Place Solution for the ECCV 2026 LifeGenIP Unlearnable Video Challenge

Paper reviews OpenAI, Anthropic, and Google agent safety incidents and proposes the PASAC framework

仅凭输出审查无法验证什么：面向研究智能体的研究契约

STCA：研究者提出针对自动驾驶视频 VLM 的时空一致性对抗攻击

POLAR：面向小型工具调用智能体的可逆性护栏框架

CCDF dataset: Grok Imagine, Veo 3.1, Sora 2 forge surveillance footage, ten deepfake detectors basically cannot recognize it

HarnessSecurity-Bench horizontally evaluates the security mechanisms of six coding agent frameworks.

论文提出 AI Agent 声明的可审计性框架

论文实测 Agent 工具调用门控栈的失败相关性

AMU：用派生血缘门控实现企业 AI Agent 列级访问控制

BabelFake：面向多语言音视频 DeepFake 检测的基准数据集

ControlPed 生成逼真行人危险动作场景，七款端到端驾驶模型 HDScore 从 88.8 降至 47.4

研究：任务范围授权可将工具调用有害执行降至零

研究测量策略网关跨平台命令裁决的能力边界

APEX: proactively defending against indirect prompt injection at the LLM agent execution boundary.

论文提出长程智能体自生子目标的运行时授权机制

论文提出 VAL 框架：相同 ASR 的 LLM Agent 防御可能对应不同安全保证

CoVer：用干预检验为 Agent 构造可判定规则边界

OmniConfess：用 Token 级「忏悔」缓解全模态幻觉

论文提出在 Kubernetes 上保护 AI Agent 的七层纵深防御框架与参考架构