Addendum to GPT-5 System Card: Sensitive conversations

OpenAI Blog Products
gpt-5 safety model-update mental-health chatgpt

Summary

OpenAI released an update to GPT-5 on October 3 to improve handling of sensitive conversations around mental and emotional distress, reducing inadequate responses by 65-80% through collaboration with 170+ mental health experts. The company published a system card addendum and safety evaluations comparing the new model to the previous August 15 version.

This system card details GPT-5’s improvements in handling sensitive conversations, including new benchmarks for emotional reliance, mental health, and jailbreak resistance.
Original Article Export to Word Export to PDF
View Cached Full Text

Cached at: 04/20/26, 02:49 PM

# Addendum to GPT-5 System Card: Sensitive conversations Source: [https://openai.com/index/gpt-5-system-card-sensitive-conversations/](https://openai.com/index/gpt-5-system-card-sensitive-conversations/) When we launched GPT‑5, we[noted⁠](https://openai.com/index/gpt-5-system-card/)in the system card that we were working to establish better benchmarks and to continue to strengthen model safety in areas related to mental and emotional distress\. On October 3, we deployed an[update⁠\(opens in a new window\)](https://help.openai.com/en/articles/9624314-model-release-notes)that reflected those efforts, improving ChatGPT’s default model to better recognize and support people in moments of distress\. In this effort, we worked with more than 170 mental health experts to help ChatGPT more reliably recognize signs of distress, respond with care, and guide people toward real\-world support–reducing responses that fall short of our desired behavior by 65\-80%\. We are publishing a related[blog post⁠](https://openai.com/index/strengthening-chatgpt-responses-in-sensitive-conversations/)that gives more information about this work, and this addendum to the GPT‑5 system card to share baseline safety evaluations\. These evaluations compare the[August 15 version⁠\(opens in a new window\)](https://help.openai.com/en/articles/6825453-chatgpt-release-notes#h_6b1adafe01)of ChatGPT’s default model, also known as GPT‑5 Instant, to the updated one launched October 3\.

Similar Articles

GPT-5.1 Instant and GPT-5.1 Thinking System Card Addendum

OpenAI Blog

OpenAI releases GPT-5.1 Instant and GPT-5.1 Thinking models with improved conversational abilities and adaptive reasoning. The system card addendum documents safety mitigations including expanded evaluations for mental health and emotional reliance.

Strengthening ChatGPT’s responses in sensitive conversations

OpenAI Blog

OpenAI has updated ChatGPT's default model to better handle sensitive mental health conversations, including improved recognition of distress, de-escalation, and routing to crisis resources. The update expands safety testing to include emotional reliance and non-suicidal mental health emergencies as standard baseline metrics.

Update to GPT-5 System Card: GPT-5.2

OpenAI Blog

OpenAI releases GPT-5.2, the latest model in the GPT-5 series, with an updated system card documenting safety mitigations and introducing GPT-5.2 Instant and GPT-5.2 Thinking variants.

GPT-5 System Card

OpenAI Blog

OpenAI releases GPT-5 System Card detailing a unified system with fast main models and deeper reasoning models routed intelligently by conversation type and complexity, featuring significant advances in hallucination reduction, instruction following, and real-world utility across writing, coding, and health domains.

Helping people when they need it most

OpenAI Blog

OpenAI shares details on ChatGPT's layered safeguards for users in mental and emotional distress, including empathetic responses, crisis hotline referrals, and human review for threats of harm to others. The post also notes GPT-5 improvements in reducing sycophancy and better handling mental health emergencies.