Black Hat Europe 2025·34 min
Patrick Ventuzelo · Nabih Benazzouz
Engineers explore architecture and security mechanisms for Breaking AI Inference Systems: Lessons From Pwn2Own Berlin.
pwn2ownbreakinginference+3
inference
34 talks tagged with this topic.
Filters+
Loading filters…
1–24 of 34 talks · loading…
Black Hat Europe 2025·28 min
Pengyu Ding · Ziteng Xu · Zhiniang Peng · Dongliang Mu
Analysts investigate attack surfaces and defenses concerning Token Injection: Crashing LLM Inference With Special Tokens.
llminjectiontoken+3
Eye on Tech·32 min
Industry experts analyze emerging threats surrounding Targeting AI: Rebellions vs. Nvidia: The Inference Chip Revolution.
rebellioninferencechip+1
IEEE Symposium on Security and Privacy·12 min
Analysts investigate attack surfaces and defenses concerning Accuracy-Privacy Trade-off in Deep Ensemble: A Membership Inference Perspective.
accuracy-privacytrade-offdeep+3
IEEE Symposium on Security and Privacy·18 min
Security researchers examine technical aspects of Automatic Inference of Search Patterns for Taint-Style Vulnerabilities.
vulnerabilitiesautomaticinference+3
IEEE Symposium on Security and Privacy·17 min
Analysts investigate attack surfaces and defenses concerning CrypTFlow : Secure TensorFlow Inference.
cryptflowtensorflowinference
IEEE Symposium on Security and Privacy·22 min
Analysts investigate attack surfaces and defenses concerning EyeTell: Video-Assisted Touchscreen Keystroke Inference from Eye Movements.
eyetellvideo-assistedtouchscreen+3
IEEE Symposium on Security and Privacy·22 min
Presenters detail vulnerability discovery and exploitation in Grammatical Inference and Language Frameworks for LANGSEC.
grammaticalinferencelanguage+3
IEEE Symposium on Security and Privacy·15 min
Engineers explore architecture and security mechanisms for Grammatical Inference and Machine Learning Approaches to Post-Hoc LangSec.
machine-learninggrammaticalinference+3
IEEE Symposium on Security and Privacy·19 min
Presenters detail vulnerability discovery and exploitation in Identifying Personal DNA Methylation Profiles by Genotype Inference.
identifyingpersonalmethylation+3
IEEE Symposium on Security and Privacy·27 min
Engineers explore architecture and security mechanisms for Membership Inference Attacks against Adversarially Robust Deep Learning Models.
deep-learningmembershipinference+2
IEEE Symposium on Security and Privacy·20 min
Security researchers examine technical aspects of Membership Inference Attacks against Machine Learning Models.
machine-learningmembershipinference+2
IEEE Symposium on Security and Privacy·22 min
Nicola
Security researchers examine technical aspects of Membership inference attacks from first principles.
membershipinferenceattack+1
IEEE Symposium on Security and Privacy·17 min
Engineers explore architecture and security mechanisms for No Pardon for the Interruption: New Inference Attacks on Android Through Interrupt Timing Analysis.
androidpardoninterruption+2
IEEE Symposium on Security and Privacy·19 min
Engineers explore architecture and security mechanisms for Property Inference from Poisoning.
propertyinferencepoisoning
IEEE Symposium on Security and Privacy·15 min
Industry experts analyze emerging threats surrounding SIRNN: A Math Library for Secure RNN Inference.
sirnnmathlibrary+1
The Linux Foundation·28 min
Aurick Qiao
Reviews arctic Inference: Open-Source LLM Inference Optimizations at Snowflake and evaluates key implications for security practitioners.
open-sourcellmarctic+2
The Linux Foundation·40 min
Michael Yuan
Examines efficient and Cross-Platform AI Inference Apps Using Rust and Wasm and evaluates key implications for security practitioners.
rustefficientcross-platform+2
The Linux Foundation·33 min
Michael Yuan · Second State
Discusses efficient and Cross-Platform LLM Inference in the Heterogenous Cloud and evaluates key implications for security practitioners.
llmefficientcross-platform+2
The Linux Foundation·23 min
Rafael Vasquez
Examines fast Inference, Furious Scaling: Leveraging VLLM With KServe and evaluates key implications for security practitioners.
fastinferencefuriou+1
The Linux Foundation·33 min
Rafael Vasquez
Examines fast Inference, Furious Scaling: Leveraging VLLM With KServe and evaluates key implications for security practitioners.
fastinferencefuriou+1
The Linux Foundation·41 min
Dalmo Cirne
Discusses fostering Trust via Explainable ML Inferences and evaluates key implications for security practitioners.
fosteringtrustexplainable+2
The Linux Foundation·37 min
Details from Hours To Milliseconds: Scaling AI Inference 10x With... Anmol Krishan Sachdeva & Paras Mamgain and evaluates key implications for security practitioners.
hourmillisecondinference+1
The Linux Foundation·34 min
Matteo Combi
Examines improve AI Inference (serving models) With KServe and VLLM and evaluates key implications for security practitioners.
improveinferenceserving+2