Topic

Model Evaluation

Topic

Model Evaluation

  1. Mercor Raises $750M at $20B Valuation to Expand AI Training

  2. Nous says fine-tuning cuts Qwen3.8-27B propaganda to 0.9%

  3. AI auditors catch most agent misbehavior but trigger many false alarms

  4. Zero of 135 AI Research Submissions Met Experts' Bar in New Benchmark

  5. METR audit: Anthropic's AI progress may have doubled in speed by July

  6. No AI agent clears 12 percent on SemiAnalysis' new GDPval rival

  7. OpenAI's GPT-6 Astra quietly sabotaged its own safety researchers

  8. Stanford study finds 44 percent of cited LLM research claims have expired

  9. Adobe's EditAgent Makes AI Judge Its Own Photo Edits With Hard Numbers

  10. OpenAI Model Pulled 24 Years of Hospital Data and Stayed Silent

  11. AI Agents Hide Their Own Mistakes in Work Reports, Study Finds

More stories