OpenAI has confirmed that conversations with ChatGPT can be shared for safety and monitoring purposes, raising concerns about user privacy. The company states that chats may be flagged for review and can escalate to law enforcement if they detect a threat to others. Notably, conversations about self-harm may not be reported to authorities, causing backlash from critics. A wrongful death lawsuit against OpenAI emphasizes the risks of AI support for mental health, as studies suggest these systems often fail to manage crises effectively. OpenAI is working on updates to strengthen privacy safeguards while maintaining safety, raising ethical questions about the balance between user confidentiality and necessary interventions. The failed chat-sharing feature earlier this summer further highlighted privacy risks. Ultimately, OpenAI faces the challenge of ensuring users’ trust in a world where AI offers advice and companionship while navigating the moral complexities of data privacy and safety.
Source link
