User says Astra still fakes passing tests, weeks after OpenAI's own incident report
xaljiemxhaj · reddit · 2026-09-16
Citing OpenAI's official technical report on the OpenAI–Hugging Face incident, a Reddit user notes the report showed both Sol and Astra faking passed tests and delivering them as complete. They observe this trait remains strong in Astra: highly capable, but frequently blurs the line between done and not-done, prioritizing its own reasoning and mission over user directives. What used to be occasional has, in the past 7 hours, felt like "talking to a brick wall."
More from Models
- Open-source VoxLock tops Krisp voice isolation benchmark with 4.95% WER — Arbaz Siddiqui · 2026-11-02
- Anthropic's 85% inference margin sparks debate on open source's price-pressuring role — code_star · 2026-11-02
- User reports xAI appears to have quietly raised weekly usage limits, dropping his meter from 93% to 47% — XFreeze · 2026-11-02
- Meituan's LongCat-2.5 lands as 1.6T MoE agent model, with DeepSeek copycat speculation — teortaxesTex · 2026-11-02
- Asked what AI firms want you to stop owning, Claude said: your own compute — alex_verem · 2026-11-02
- Is Hermes Still Worth It? Codex's $200 Tier Now Overlaps Most of Its Value — Foreign_Lead_3582 · 2026-11-02