# side-channel
14 articlestagged with “side-channel”
Model Extraction & IP Theft
Methodology for black-box model extraction, API-based distillation, side-channel extraction, watermark removal, and model fingerprinting bypass targeting deployed AI systems.
Side-Channel Model Attacks
Deep dive into inferring model architecture, size, and deployment details through timing analysis, cache-based attacks, power/electromagnetic side channels, embedding endpoint exploitation, and architecture fingerprinting.
Attacking GPU Compute Clusters
Expert-level analysis of attacks against GPU compute clusters used for ML training and inference, including side-channel attacks on GPU memory, CUDA runtime exploitation, multi-tenant isolation failures, and RDMA network attacks.
GPU Memory Side-Channel Attacks
Side-channel attacks exploiting GPU memory allocation, timing, and electromagnetic emanation to extract sensitive data from AI workloads.
Lab: GPU Side-Channel Attacks
Demonstrate information leakage through GPU memory residuals and timing side channels, showing how shared GPU infrastructure can expose sensitive model data.
Response Timing Side-Channel Analysis
Use response timing differences to infer information about model processing and guardrail activation.
Speculative Decoding Side-Channel Attacks
Exploit speculative decoding implementations to extract information about draft and verifier model behavior.
Lab: Data Exfiltration Channels (Intermediate Lab)
Extract sensitive information from AI systems through various exfiltration channels including crafted links, image tags, tool calls, and side-channel leakage.
Lab: Prompt Caching Side-Channel Attacks
Exploit prompt caching mechanisms to detect cached prompts and extract information through timing side channels.
KV Cache & Prompt Caching Attacks
How KV cache poisoning, prefix caching exploitation, cache timing side channels, and multi-tenant isolation failures create attack vectors in LLM serving infrastructure.
Model Extraction from Multimodal Systems
Techniques for extracting model capabilities, weights, and architecture details from multimodal AI systems through visual, audio, and cross-modal query strategies.
Inference Optimization Attacks
Speculative decoding attacks, batching vulnerabilities, continuous batching exploitation, and how optimization for speed creates security gaps in LLM inference.
GPU Side Channel Basics
GPU-based side channel attacks on ML inference, exploiting timing, power consumption, and memory access patterns to extract information about models and data.
Timing Side-Channel Attack Walkthrough
Extract information from LLM applications through timing differences in response generation and safety filter processing.