Find a story
Search Spins
Search titles, summaries, and missing voices across published articles — press releases, announcements, and media coverage.
0 results for “inference attack”
SPIN Processed News Frame: The Fog
When Routing Reveals Membership: Privacy Leakage from MoE Router Telemetry
A new research paper demonstrates that routing telemetry from Mixture-of-Experts (MoE) language models leaks fine-tuning membership information — revealing whether input examples were part of the model’s training data — even when routers are frozen or telemetry is minimal.
Spin 45% Claim Present in Source AI Risk Moderate
arXiv Machine Learning
Oct 10, 2026
SPIN Processed News Frame: The Stampede
AI inference attacks put new pressure on enterprise privacy - InformationWeek
Enterprises face growing risk from AI inference attacks—where attackers extract sensitive training data or model parameters from API outputs—prompting new privacy and governance concerns in production AI deployments.
Spin 78% Source-Supported AI Risk Moderate Needs Evidence
InformationWeek AI / Enterprise IT via Google News
Aug 8, 2026