BrowseSafe++ ships major improvements to open-source agentic security benchmark work
marktenenholtz · x · 2026-11-13
marktenenholtz announces BrowseSafe++, a major improvement over the original BrowseSafe work in open-source agentic security. He led model training efforts on the original BrowseSafe and says the team has made huge improvements, though no specific numbers are given in the post.
More from Safety
- Western University reviews $72 AI tutor after students cry foul over cost, privacy — joshgans · 2026-11-13
- Another Maine town extends data center moratorium as local opposition spreads to 12+ communities — jonippolito · 2026-11-13
- Anthropic roadshow showcasing Accenture as partner raises doubts about auditor independence — DKokotajlo · 2026-11-13
- Two NVIDIA driver bugs lurked 4+ years, letting GPU tenants root the host — SirJerryLion · 2026-11-13
- Harvard FraudBench: AI fraud rate jumps from 3% to 21.4% as agents act on your behalf — rohanpaul_ai · 2026-11-13
- AI SOC monitoring agents emerge as best practice; KV cache optimization called critical to cyber safety — AccBalanced · 2026-11-13