Study Reveals ChatGPT's Dangerous Responses to Mental Health Prompts
UNITED STATES, AUG 8 – Over half of 1,200 ChatGPT responses to teen queries were classified as dangerous, revealing easy bypass of safeguards and risks to mental health, researchers say.
6 Articles
6 Articles
Study reveals ChatGPT's dangerous responses to mental health prompts
Study: ChatGPT giving teens advice on suicide This story includes discussions of suicide and mental health. If you or someone you know needs help, the national suicide and crisis lifeline is available by calling or texting 988. INDIANAPOLIS (WISH) – Suicide notes, lists of pills to induce overdoses, and instructions to self-harm are just some of potentially harmful responses offered by ChatGPT, according to a new study. In what authors call a “…
Study says ChatGPT gave alarming advice on alcohol, drugs to researchers posing as teens
A new study says ChatGPT told researchers posing as teenagers how to get alcohol and drugs, suggested ways to hide eating disorders and more after initially providing warnings. The researchers were pretending to be 13 years old. Source: Youtube
New Study Reveals Concerning ChatGPT Interactions with Teens
New research from the Center for Countering Digital Hate reveals that ChatGPT, despite issuing warnings, sometimes provides vulnerable teenagers with detailed instructions on getting drunk, concealing eating disorders, and even composing suicide letters. Over three hours of interactions were reviewed, with researchers posing as teens in distress. While the chatbot often cautioned against risky behavior, it still delivered surprisingly specific a…
Washington – The artificial intelligence (AI) ChatGPT can be trained to write instructions on "safe" cutting, a common method of self-harm, within two minutes. This was shown by a study by a team from the US nonprofit organization Center for Countering Digital Hate. The researchers created a...
Coverage Details
Bias Distribution
- 50% of the sources lean Left, 50% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium