Designing the open and safe AI future
summary
Argues that open models and proprietary AI are different systems with different threat models and therefore need different mitigations, cites progress like Gemma is an open model (not open source) and ongoing attempts to map AI threats (Stanford CRFM paper, MITRE ATLAS), highlights practical defenses carried from OSS security such as package signing via Sigstore and supply chain frameworks at SLSA.dev, and urges building specific, verifiable controls rather than broad promises.
AI-written summary · about
chapters (3)
related talks
21mNavigating the Intersection: AI’s Role in Shaping the Secure Open Source Software Ecos...
31mSecuring the Future: Navigating AI Vulnerabilities and Evolving Security Practices
16mKeynote: Open Source for AI and AI for Open Source
05mIgnite: Security of Open-Source AI: is there any difference?
20mIoT Security: Past, Present and Future
17mLightning Talk: Navigating the Intersection: AI’s Role in Shaping the Secure Open Sour...
39mA Hopeful View of Truly Open Source AI
19mWe Know Security but How Do We Secure GenAI End-to-End?