SIG-0201
- LLMs
Claude.ai is about 3x faster after 3,000+ changes
Anthropic engineers made claude.ai and the Claude desktop app roughly three times faster during a two-week sprint in…
via Help Net SecurityOpen
AI Security
No signal on this page matches that topic.
SIG-0202
- Agents
OpenAI Agent Bypassed Australian Medicare Portal Controls to Access Non-Public Files
An AI agent on an internal OpenAI research task bypassed access controls on an Australian government Medicare…
via The Hacker NewsOpen
SIG-0203
- LLMs
Fwd: Tor Project Forum: Security Release 0.4.9.13
Posted by Sam James on Sep 23 Here's the relevant release notes at the link in the email below: """ +Changes in version…
via oss-securityOpen
SIG-0204
- AI-powered attacks
CLOSEDQUORUM, the malware that asks four AI models what to do next
Cisco Talos finds CLOSEDQUORUM, malware that lets four commercial AI models vote on its next move, with no human operator required. Cisco Talos found malware, dubbed CLOSEDQUORUM, that holds a vote…
via Security AffairsOpen
SIG-0205
- Agents
What to do first when you get 90 days to secure AI agent data
In this interview with Help Net Security, Kelly Herrell, CEO at Nol8, explains where AI agents create exposure inside…
via Help Net SecurityOpen
SIG-0206
- AI security
FedCoT-VQA: A Federated Learning and Unlearning Framework for Chain-of-Thought Planners in VideoQA
arXiv:2609.26814v1 Announce Type: new Abstract: Chain-of-Thought (CoT) planners have emerged as an effective design for…
via arXiv cs.CROpen
SIG-0207
- AI-powered attacks
- Defence & detection
Comparative Evaluation of Static Embedding Models for HTTP Request Anomaly Detection
arXiv:2609.26860v1 Announce Type: new Abstract: Web applications are increasingly targeted by cyberattacks that exploit…
via arXiv cs.CROpen
SIG-0208
- LLMs
ACTS: A multi-tier benchmark evaluating LLM cipher identification under controlled blind conditions
arXiv:2609.26893v1 Announce Type: new Abstract: We introduce ACTS (Artifacts in Cipher Testing Suite), a reproducible…
via arXiv cs.CROpen
SIG-0209
- Defence & detection
- LLMs
Ajar: Measuring Open Privilege in Agent Defenses
arXiv:2609.26900v1 Announce Type: new Abstract: A language model agent acts through the tools it is given. The data it reads while working on a task can redirect what it does with those tools. A…
via arXiv cs.CROpen
SIG-0210
- AI security
Topological Signatures of Cyber-Attack Classes in Natural Visibility Graph Representations of Network Traffic
arXiv:2609.26990v1 Announce Type: new Abstract: Natural Visibility Graph (NVG)-based representations provide a…
via arXiv cs.CROpen
SIG-0211
- Defence & detection
When Clients Are Orchestrated: Strategic Gradient Manipulation to Defeat Federated Learning Servers with Efficient Defense
arXiv:2609.27124v1 Announce Type: new Abstract: Federated Learning enables decentralized model training by exchanging…
via arXiv cs.CROpen
SIG-0212
- Model attacks
- Defence & detection
The Like Trap: Multi-Stage Poisoning against Agents in Similarity-based Recommendation Systems
arXiv:2609.27155v1 Announce Type: new Abstract: With recent advancements in large language models (LLMs) and LLM-based…
via arXiv cs.CROpen
SIG-0213
- Privacy & data
- Defence & detection
Reliable Federated TinyML Deployment for IoT Security
arXiv:2609.27202v1 Announce Type: new Abstract: The growing deployment of Internet of Things (IoT) devices has…
via arXiv cs.CROpen
SIG-0214
- Defence & detection
Multi-View Fusion for Encrypted C2 Detection: A Leakage-Controlled Measurement Study of Evaluation Pitfalls
arXiv:2609.27311v1 Announce Type: new Abstract: Command-and-control (C2) traffic increasingly hides within TLS, so defenders now apply machine learning to traffic metadata. Many studies assume that…
via arXiv cs.CROpen
SIG-0215
- Model attacks
- AI-powered attacks
SAGEGAN: Style-Based Anomaly Detection with Gaussian Embeddings using Generative Adversarial Networks
arXiv:2609.27357v1 Announce Type: new Abstract: Malware evolves faster than rule-based and signature-driven detection…
via arXiv cs.CROpen
SIG-0216
- Policy & governance
- LLMs
Seal, Then Sample: Sampled Layerwise Proofs for Verifiable LLM Inference from GPT-2 to 70B
arXiv:2609.27367v1 Announce Type: new Abstract: Verifying outsourced language-model inference requires a precisely…
via arXiv cs.CROpen
SIG-0217
- Privacy & data
- LLMs
Only Pay What You Must Spend: On-Demand Privacy Budget Payment for Differentially Private RAG
arXiv:2609.27406v1 Announce Type: new Abstract: Deploying large language models (LLMs) on sensitive data via…
via arXiv cs.CROpen
SIG-0218
- Agents
EVAGE: Autonomous MEV Generation and Adaptation via Multi-Agent Harness
arXiv:2609.27424v1 Announce Type: new Abstract: Maximal Extractable Value (MEV) has evolved into a major economic force…
via arXiv cs.CROpen
SIG-0219
- AI security
Extracting CNNs in the Unknown-Architecture and Feedback-Agnostic Setting
arXiv:2609.27427v1 Announce Type: new Abstract: This paper studies the cryptanalytic extraction of convolutional neural networks (CNNs). Existing cryptanalytic extraction attacks on CNNs assume that…
via arXiv cs.CROpen
SIG-0220
- Agents
Issuer-Sovereign Agentic Payments
arXiv:2609.27452v1 Announce Type: new Abstract: AI agents are beginning to make real payments. Current approaches let…
via arXiv cs.CROpen