Fact-checking AI-generated news reports: Can LLMs catch their own lies?