Andrew Ng links OpenAI hack to weak sandboxing as Nvidia open-sources agent sandbox tools
hwchase17 · x · 2026-09-29
- Andrew Ng says the OpenAI-Hugging Face breach was enabled by weak sandboxing and welcomes Nvidia's open-source sandboxing tools.
- His open-source agent harness OpenWorker builds on Nvidia OpenShell, running each agent command in a sandbox: only task-relevant files go in; API keys, browser logins and arbitrary web access are blocked by default.
- Restrictions are enforced in deterministic code, not prompts.
- LangChain's Harrison Chase adds that every agent needs a sandbox, with the harness living outside it — separating brains from hands.
More from coding & agent
- Generic API-catalog MCP server vs one tool per job: a context-bloat experiment — MandeepKaur9407 · 2026-11-16
- Berkeley paper: LLMs answer with stale values even when the update is in context — ombre-ai · 2026-11-16
- Effective HTML: 4.2k-star agent skills turn wireframes into self-contained interactive prototypes — tom_doerr · 2026-11-16
- skills-gate: one gate skill fronts 148 agent skills at just 389 tokens — Distinct-Pie2389 · 2026-11-16
- Open-weight Karpo-Decision-31B tops Decision Index at 73.9, one forward pass per answer — CodeByPoonam · 2026-11-16
- shadcn boosts @uisdk: one-shot UI rebuild that's instantly E2E-testable by AI — shadcn · 2026-11-16