How EuroSafeGuard works: activation probes triage traffic before heavy classifiers
ZhijingJin · x · 2026-11-07
EuroSafeGuard is a three-stage cascade: a cheap activation probe reads the protected model's hidden states and scores every request; suspicious ones go to a fine-tuned input classifier (Qwen3.5-9B); an exchange classifier then scores the request together with the model's reply. A weighted sum decides whether to show or block the reply. Built by EuroSafeAI with the University of Toronto & Vector Institute and Max Planck Institute, it also ships EuroSafe-CC++, an open recreation of Anthropic's cascade.
Related event: EuroSafeGuard Tops 16 Open Guardrails With 93% Jailbreak Detection(3 posts)→
More from Safety
- Nigerian-American scientist Chinasa T. Okolo named to TIME100 AI list — ChinasaTOkolo · 2026-11-07
- Ex-OpenAI policy head: safety reports yield diminishing returns without human-level framing — deanwball · 2026-11-07
- OpenAI employee pleads not guilty to secretly bugging company office — Polymarket · 2026-11-07
- California bill would force frontier AI labs to disclose compute allocation for safety — ohlennart · 2026-11-07
- Critic: Anthropic backs recursive self-improvement at full speed while only talking about safety — peterwildeford · 2026-11-07
- EuroSafeGuard adds only 1% compute overhead for full-traffic jailbreak defense — ZhijingJin · 2026-11-07