AI Threat AlertOn the Fragility of Contribution Score Computation in Federated Learning
alter the final scores. Second, we explore vulnerabilities posed by poisoning attacks, where malicious participants strategically manipulate their model updates to inflate their own contribution scores or reduce the importance
Targeting World Models to Compromise Robot Learning Pipelines
While highly practical, in this work we demonstrate that world models introduce a uniquely stealthy and effective data poisoning entry point into the robot learning supply chain that can result
BadRSSD: Backdoor Attacks on Regularized Self-Supervised Diffusion Models
backdoor attack targeting the representation layer of self-supervised diffusion models. Specifically, it hijacks the semantic representations of poisoned samples with triggers in Principal Component Analysis (PCA) space toward those
IssueTrojanBench: Benchmarking AI Coding Agents Against Malicious Issue Requests
security risks from both the LLM backbone, where adversarial prompts, poisoned training data, and backdoor triggers can cause models to emit insecure or attacker-chosen code, and their agentic architecture
Backdoor Containment via Expert Quarantine and Shutdown in LLMs
regularization-steered MoE-like setting. Specifically, given a poisoned dataset, QES augments a Transformer-based language model with routed expert-specific LoRA branches and lightweight routers, and uses auxiliary routing
Memory Poisoning Attack and Defense on Memory Based LLM-Agents
Large language model agents equipped with persistent memory are vulnerable to memory poisoning attacks, where adversaries inject malicious instructions through query only interactions that corrupt the agents long term memory
Backdoor Attribution: Elucidating and Controlling Backdoor in Language Models
Fine-tuned Large Language Models (LLMs) are vulnerable to backdoor attacks through data poisoning, yet the internal mechanisms governing these attacks remain a black box. Previous research on interpretability
TRIS: A Tri-Layer Retrieval Integrity Sieve Against Knowledge Poisoning
with an independent judge model, structural filtering of trigger-payload artifacts, and LLM consistency verification. The design exploits a key weakness of retrieval-stage poisoning: a single document must satisfy
Anti-Backdoor Coreset Selection via Cumulative Entropy
time defenses against neural backdoors isolate a benign subset from poisoned training data, to learn a backdoor-free model from it. In this paper, we formulate this defense strategy
The Framing Gap: Indirect Prompt-Injection Exfiltration Defeats Surface-Level Defenses in Tool-Using Agents
safe synthetic lab (canary secret, mock tools, matched clean-vs-poisoned metric) we report the framing gap: across six models, ten overt injection classes are refused
Assessing the Operational Impact of Poisoning Attacks over Augmented 3D Point Cloud Public Datasets for Connected and Autonomous Vehicles
system apply over the learned models. Its impact over data augmentation models is unclear. While data augmentation reduces the likelihood of poisoning attack success, some valid questions remain. Is data
Poisoned Identifiers Survive LLM Deobfuscation: A Case Study on Claude Opus 4.6
poisoned identifier names in the string table survive into the model's reconstructed code, even when the model demonstrably understands the correct semantics? Using Claude Opus 4.6 across 192 inference
RefineRAG: Word-Level Poisoning Attacks via Retriever-Guided Text Refinement
Retrieval-Augmented Generation (RAG) significantly enhances Large Language Models (LLMs), but simultaneously exposes a critical vulnerability to knowledge poisoning attacks. Existing attack methods like PoisonedRAG remain detectable due to coarse
Cordon-MAS: Defending RAG against Knowledge Poisoning via Information-Flow Control
Confundo-style poisoning where adversarially optimized documents manipulate generated outputs. Existing defenses assume that detecting poisoned evidence prevents harm. We show this assumption is incorrect: models exhibit a monitoring-control
Phantom Transfer: Data-level Defences are Insufficient Against Data Poisoning
data-level defences are insufficient for stopping sophisticated data poisoning attacks. We suggest that future work should focus on model audits and white-box security methods
TRUSTDESC: Preventing Tool Poisoning in LLM Applications via Trusted Description Generation
injection attack surface: tool poisoning attacks (TPAs). Attackers manipulate tool descriptions by embedding malicious instructions (explicit TPAs) or misleading claims (implicit TPAs) to influence model behavior and tool selection. Existing
ToxicRAG: Compromising Retrieval-Augmented Generation Systems via Single-Shot Knowledge Poisoning Attacks
Augmented Generation (RAG) can ground large language model (LLM) outputs in external evidence, but it also exposes the system to knowledge poisoning. Representative attacks use multiple injected documents or templates
Trustworthy Blockchain-based Federated Learning for Electronic Health Records: Securing Participant Identity with Decentralized Identifiers and Verifiable Credentials
patient data. Despite its potential, FL remains vulnerable to poisoning and Sybil attacks, in which malicious participants corrupt the global model or infiltrate the network using fake identities. While recent
CIBER: A Comprehensive Benchmark for Security Evaluation of Code Interpreter Agents
four major types of adversarial attacks: Direct/Indirect Prompt Injection, Memory Poisoning, and Prompt-based Backdoor. We evaluate six foundation models across two representative code interpreter agents (OpenInterpreter and OpenCodeInterpreter), incorporating
The safety failures we are not instrumenting: a perspective on hidden safety-critical challenges in modern AI systems
legitimacy laundering in retrieval, prompt injection, reward hacking, memory poisoning, evaluation deception, fictional human oversight, synthetic evidence pollution, and model collapse. We conclude with design and governance recommendations