"Elon Musk's AI chatbot blames safeguard 'lapses' over child sexual images" - Original FT headline

"Musk's Grok AI bot is fixing safeguard lapses after posting of sexualized images of Children - Original headline

Both were fixed, but these articles suggest LLM was 1. providing its own answers to questions, 2. able to fix them

Headlines were fixed but quite revealing how these systems are nonsensical ways to escape responsibility