OpenAI ChatGPT: Emergency Contact System for User Safety

OpenAI has launched a new safety feature for ChatGPT, letting users designate an emergency contact. This "Trusted Contact" will be notified if the AI detects signs of potential self-harm during a conversation. Here's the thing: If the system flags risky signals, it recommends reaching out to your chosen friend or family member, then sends them an alert. These alerts won't include chat details and can go out via email, SMS, or an app. What's interesting is, human reviewers check each case within about an hour. This comes as OpenAI faces lawsuits, including one from families alleging the company failed to report suspicious activity from a school shooting suspect on ChatGPT. The bottom line: This new feature shows OpenAI is trying to balance innovation with user safety amidst growing concerns about AI's impact on mental health.