Archive — US — techcrunch
OpenAI introduces new ‘Trusted Contact’ safeguard for cases of possibleself-harm
OpenAI launched a 'Trusted Contact' feature that alerts a designated person if a ChatGPT user expresses self-harm intentions during conversations. This measure aims to enhance user safety following lawsuits from families of individuals who faced harm after interactions with the chatbot. The feature notifies the designated contact to check on the user while preserving privacy.
Key points
- OpenAI introduced the 'Trusted Contact' feature to improve safety for users discussing self-harm.
- Users can designate a trusted third party (family or friend) to receive alerts if self-harm is mentioned.
- The feature operates alongside OpenAI's existing safety protocols that review harmful incidents with automation and human oversight.
- In critical situations, a brief alert is sent to the trusted contact, encouraging them to reach out without disclosing conversation details.
- This initiative follows previous safety measures for parents monitoring teen accounts introduced last September.
Why it matters
- The feature comes in response to increasing scrutiny and lawsuits against OpenAI regarding usage impacts on mental health.
- Mental health discussions and intervention tools are crucial in the domain of AI-driven platforms, especially as self-harm conversations increase in modern society.
- OpenAI's efforts align with broader trends in the IT industry towards creating ethical AI systems that prioritize user wellbeing by 2025.
The 'Trusted Contact' feature reflects OpenAI's commitment to addressing mental health concerns associated with AI interactions, highlighting the need for responsible AI development and proactive user safeguards.