Topic: content moderation

  • Google's Gemini AI Model Shows Safety Decline

    Google's Gemini AI Model Shows Safety Decline

    Google's Gemini 2.5 Flash AI model shows a **4.1% decline in text-to-text safety** and a **9.6% drop in image-to-text safety** compared to its predecessor, raising concerns about content moderation. Major AI developers like Google, Meta, and OpenAI are prioritizing **permissiveness**, but this sh...

    Read More »
  • Wikipedia Embraces AI While Keeping Human Volunteers

    Wikipedia Embraces AI While Keeping Human Volunteers

    Wikipedia is integrating AI as an assistant to support human contributors, not replace them, with a three-year strategy focused on collaboration. AI will automate repetitive tasks like translations and onboarding, allowing editors to focus on curating accurate content through human oversight. The...

    Read More »