firasd

Recent Submissions

1. 1. ▲ Strangerbench: A benchmark for AI forecasting after training cut-off dates (github.com) 1 point · 1 hour ago · 0 comments
2. 2. ▲ Sam Altman's blind spot on AI model power (vibesbench.substack.com) 3 points · 11 days ago · 1 comment
3. 3. ▲ Dullness and Disbelief: The 2026 AI Regression (vibesbench.substack.com) 2 points · 16 days ago · 0 comments
4. 4. ▲ Claude output silently rewritten by Anthropic (github.com) 2 points · 20 days ago · 0 comments
5. 5. ▲ Anthropic silently rewriting Claude punctuation output in API (github.com) 2 points · 22 days ago · 1 comment
6. 6. ▲ AI sycophancy panic (github.com) 57 points · 25 days ago · 102 comments
7. 7. ▲ Vibesbench: Prompts to track conversational regression in AI models (github.com) 3 points · 28 days ago · 0 comments
8. 8. ▲ Stockfish shows Morphy and Fischer didn't sacrifice their queens (github.com) 1 point · 1 month ago · 1 comment
9. 9. ▲ Vibesbench: A Multi-Turn Conversational AI Benchmark (github.com) 1 point · 1 month ago · 0 comments