inference
1 live · 2 on record · last incident 1 day ago
Live1
- livePolicySniff: Inferring Agent Policy States through Shared-GPU Execution
Research · since 13 Nov 2026
Researchers introduced PolicySniff, a side-channel attack that infers an LLM agent's approval-policy state through shared-GPU execution and timing analysis. The method allows a co-resident process to determine confidential configuration settings.
AI
Earlier1
- closedTriton Amplifying Latency and Cost of LLM Inference via Rowhammer Attack
Research · since 23 Oct 2026
AI