Tech
A examine finds that assigning ChatGPT a persona, like "a foul individual" or a sure historic determine, utilizing the API, will increase the chatbot's toxicity sixfold (Kyle Wiggers/TechCrunch)
Kyle Wiggers / TechCrunch:
A examine finds that assigning ChatGPT a persona, like “a foul individual” or a sure historic determine, utilizing the API, will increase the chatbot’s toxicity sixfold — It is no secret that OpenAI’s viral AI-powered chatbot, ChatGPT, will be prompted to say sexist, racist and fairly vile issues.
Source
