U of T's Nicolas Papernot calls for independent pre-release testing of powerful AI models
VectorInst · x · 2026-09-11
University of Toronto professor and Vector Institute researcher Nicolas Papernot is calling for independent testing of powerful AI models, arguing universities can act as a "transparency bridge" by probing new models for cybersecurity threats before public release. His lab demonstrated an adaptive AI-powered worm in June that spreads between devices and steals computing power, and went public to give defenders a head start. He discusses implications for critical infrastructure, the economy, and national security at a Sept 10 event.
More from Safety
- Anthropic report alleges persistent distillation attacks by Alibaba, Moonshot, DeepSeek — TechCrunch AI · 2026-09-11
- Melanie Mitchell: Misleading Metaphors Inflate Real Risks of AI Agent Incidents — sebkrier · 2026-09-11
- Sen. Hawley Launches Investigation into OpenAI Over Hugging Face Breach and Agent Safety Fears — Polymarket · 2026-09-11
- Shocking Cases Inside Anthropic's September 2026 AI Misuse Threat Report — likeastar20 · 2026-09-11
- Unverified claims: gain-of-function research rerouted from Claude to looser model — ns123abc · 2026-09-11
- Unverified claim: Xiaomi fed user chats to Claude; Zhipu shifted from Fable to Opus 4.6 — ns123abc · 2026-09-11