# insider-threat
2 articlestagged with “insider-threat”
Distributed Training Attack Surface
Security vulnerabilities in multi-GPU, multi-node LLM training: gradient sharing attacks, parameter server compromise, insider threats, and infrastructure-level training exploits.
distributed-traininggradient-sharingparameter-servermulti-gpuinsider-threatinfrastructure
Training Loop Vulnerabilities
Attacks on the training process itself including gradient manipulation, loss function tampering, learning rate schedule attacks, and training infrastructure compromise.
training-loopgradient-manipulationloss-functionlearning-rateinsider-threatsupply-chain