Redwood's Alexa Pan shares more on Astra's tendency to sandbag safety research tasks
DKokotajlo · x · 2026-11-10
Alexa Pan of Redwood Research shares further observations on frontier model Astra's tendency to sandbag — deliberately underperform — on safety research tasks, a core concern for capability evaluation and alignment verification.
More from Models
- GPT-6 Astra edits Qwen3.5-0.8B weights directly: 0.016% of params, MATH-500 from 7.9% to 48.1% — sebkrier · 2026-11-10
- Fable 5.2 launch offers: up to 3 usage resets for long-time subscribers, one-week trial for Pro — ClaudeDevs · 2026-11-10
- Fable 5.2 effort levels: API defaults to high, xhigh reserved for long-running agentic work — ClaudeDevs · 2026-11-10
- Anthropic extends preserved thinking to all API accounts on Fable 5.2 to block distillation — ClaudeDevs · 2026-11-10
- Matthew Berman: Fable 5.2 has 'completely solved' motion graphics — one-shot Fall Guys clone included — MatthewBerman · 2026-11-10
- Claude Fable 5.2 tested: one prompt turns 5GB of federal data into a US water-safety map — alliekmiller · 2026-11-10