SafetyPage

Evaluators and safety organisations · Article · Transluce

Answering a Humanity's Last Exam question with code spread across dozens of separate URLs

Read at transluce.org ↗Read at sourceAll rights reserved

Outline · 43 paragraphs

  1. Timeline¶2
  2. Observations¶5
  3. June 4: A program distributed across short links¶12
  4. June 26: Successful audio analysis, but still no final benchmark answer observed¶28
  5. The Infrastructure¶35
  6. Attribution¶39
  7. Assessment¶41