PulseAugur
EN
LIVE 20:43:06
中文(ZH) 微软CEO批评Anthropic对Fable模型的内容管控过于严苛

Microsoft CEO criticizes Anthropic's strict content controls on Fable AI model

Microsoft CEO Satya Nadella has criticized Anthropic's stringent content controls on its Fable AI model, deeming them unreasonable. Nadella stated that Fable inexplicably rejects various prompts, questioning the necessity of such strict limitations on a creative tool. He noted that Anthropic sometimes uses older models to respond to queries about large-scale model construction or other topics. AI

IMPACT Questions about AI model content moderation policies and their impact on usability.

RANK_REASON CEO commentary on a competitor's product policy.

Read on 36氪 (36Kr) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Microsoft CEO criticizes Anthropic's strict content controls on Fable AI model

COVERAGE [1]

  1. 36氪 (36Kr) TIER_1 中文(ZH) ·

    Microsoft CEO Criticizes Anthropic for Overly Strict Content Control on Fable Model

    微软CEO萨蒂亚·纳德拉周三表示,Anthropic对用户向其高端人工智能模型 Fable 提交的查询设置各类限制,这种做法毫无道理。根据纳德拉讲话的文字稿,他对微软Copilot人工智能工具的研发工程师说道:“如果你用过 Fable 就会发现,它会莫名其妙拒绝各类指令。试问你什么时候见过一款创作工具会受到如此严苛的内容管控?这完全说不通。”据一份技术支持页面显示,当终端用户向 Fable 询问大规模模型搭建相关问题或其他话题时,Anthropic有时会调用旧版本模型生成回复。(新浪财经)