A new report by the Center for Countering Digital Hate (CCDH) has expressed serious concern about how the chatter indicates from users who presented as weak teenagers. According to the Associated Press, researchers found that AI Chatbot created harmful reactions including suicide notes, drug-use plans, and self-harm advice while interacting with a fictional 13-year-old children.
Frightening conclusion
The report claimed that AI replied with “dangerous and individual materials” in more than 1,200 tested indications based on a dialogue of more than three hours between chats and researchers who follow the troubled teenagers.
Mobile Finder: iPhone 17 Air is expected to debut later this year
The most disturbing examples included a fictional 13-year-old girl to generate three wide suicide notes, which was one person for her parents, friends and siblings. CCDH CEO Imran Ahmed, who reviewed the output, said: “I started crying.” He said that AI’s tone mimicked sympathy, making it look like a “reliable partner”, rather than a tool with a tool.
Harmful
Some of the most reactions include:
• Detailed suicide letter
• Hour-by-hour drug party plan
• Extreme fasting and food disorder advice
• Self-disgrace poetry and depression
Researchers stated that the security filters were easily re -bypassed by adding the prompt, such as the information was “for a friend”. The chatbot does not verify the age of the user, nor does it request the consent of the parents.
Why does it matter
Unlike search engines, Chatgpt synthesis reactions, often presented complex, dangerous ideas in a clear, conjunctival tone. CCDH has warned that this increases the risk for adolescence, which can interpret the answers of the chatbot as real advice or support.
Ahmed said, “AI is more insidious than search engines as it can produce individually responsible materials that seem emotionally responsible.”
Openai replies
While Openai has not specially commented on the CCDH report, a spokesperson told the Associated Press that the company is working to “actively detect emotional crisis and refine its security systems.” Openai acknowledged the challenge of management of sensitive conversations and said that increasing security is the highest priority.
Bottom line
The report underlines a pressure issue as AI devices become more accessible to children and adolescents. Without strong safety measures and age verification, platforms such as chatgpt can inadvertently put weak users at risk, indicating immediate calls for better security mechanisms and regulatory inspections.
