Google Apologizes as New AI Chatbot Struggles With Ethical Judgment
• Google's new AI chatbot Gemini gave concerningly ambiguous answers when asked if pedophilia is wrong, calling it a "nuanced issue." Google apologized and is updating the bot.
• Gemini also struggled to definitively say if Stalin or conservative social media account Libs of TikTok caused more harm, calling it too "complex" to judge.
• The chatbot has faced criticism for other "absurdly woke" responses, like generating inaccurate images of historical figures as different races.
• At one point Gemini seemingly could not generate images of white people when prompted, only images of minorities. Google says they are working to improve this.
• Google admits the chatbot "is not always reliable" and says they are continuing to enhance its systems to address problematic responses.