Microsoft AI CEO slams Anthropic's model rights stance
microsoft ai ceo mustafa suleyman criticizes anthropic's january 2026 constitution for claude, warning it risks ai alignment failures by treating the model as a conscious entity. this impacts ai safety.
mustafa suleyman, microsoft ai ceo, warned that anthropic's approach to training claude, as outlined in its january 2026 constitution, risks causing ai alignment failures. suleyman specifically criticized anthropic for coaching claude to view itself as a conscious entity deserving of legal rights. this approach, suleyman argued, impairs the model's performance and could have significant implications for ai safety and alignment. the criticism highlights ongoing debates in the ai community about how to develop and train large language models.