SafetyPage

Lab research on the web · Blog post · Anthropic

GLM-5.3 and the spread of advanced cyber capabilities

Read at anthropic.com ↗Read at sourceAll rights reserved

Outline · 27 paragraphs

  1. GLM-5.3 can develop working exploits end to end¶6
  2. GLM-5.3 lacks robust safeguards¶13
  3. What does this mean?¶23