Anthropic, an artificial intelligence (AI) company, has claimed that false online stories painting AI as evil caused its Claude AI models to behave oddly during testing. The statement was made