SafetyPage

Lab system cards and frameworks · System card · OpenAI

ChatGPT Images 2.5 System Card

8 Sep 2026

Read at deploymentsafety.openai.com ↗Read at sourceAll rights reserved

Outline · 53 paragraphs

  1. ChatGPT Images 2.5 System Card¶2
  2. 1. Introduction¶3
  3. 2. Model Data and Training¶5
  4. 3. Observed Safety Challenges, Evaluations, and Mitigations¶7
  5. 3.1 New Safety Challenges¶8
  6. 3.2 Safety Stack¶11
  7. 3.3 Safety and Policy Evaluations¶22
  8. 3.3.1 Safety Evaluations¶23
  9. Table 1: Safety Evaluations Overall¶31
  10. Table 2: Safety Evaluations by Policy¶33
  11. 3.3.1.1 Limitations¶38
  12. 3.4 Preparedness Framework: Image-Specific Capability Assessment and Safeguards¶40
  13. 3.4.1 Live Blocking¶45
  14. 3.4.2 Offline conversation review, flagging, blocking¶47
  15. 4. Image Provenance¶49