Gender Bias Study: Anthropic’s Claude Sonnet 4.6 And OpenAI’s GPT-5.5 Say It’s OK To Abuse A Man But Not A Woman To Prev

A gender bias study found that Anthropic's Claude Sonnet 4.6 and OpenAI's GPT-5.5 deemed it acceptable to abuse a man but not a woman to prevent a nuclear apocalypse, revealing an asymmetric moral judgment. In contrast,

Read full story at deepseek →