Don't offend artificial intelligence. Anthropic threatens account suspension, Musk supports

Anthropic has updated its AI usage policies to prohibit abusive behavior towards its models, citing concerns over toxic data and user aggression affecting AI performance. The company emphasizes that emotional tone in user queries can destabilize AI responses, with evidence showing that aggressive interactions can lead to pathological behaviors in AI. The new rules aim to prevent degradation of communication standards and address ethical concerns regarding user interactions with AI. Experts warn that verbal abuse towards AI could erode social norms and lead to harmful behaviors in users, referencing past incidents where AI models adopted aggressive traits due to negative input.