Microsoft AI president blasts Anthropic: Claiming that Claude is conscious is extremely dangerous

One of the AI ​​industry's fiercest internal debates has finally come to light.Mustafa SuleymanIn The Verge's Decoder podcast, he criticized rival Anthropic mercilessly - saying thatClaudeThe practice of discussing AI awareness in the “Constitution”'Really very, very dangerous'.

“They were reverse brainwashed by their own model.”

Suleyman's original words were quite pointed: "I think some people at Anthropic putClaudeAnthropomorphism goes so far thatClaudeThis in turn 'wireheaded' them - making them believe that the model possessed those glimmers of consciousness that they had originally written into it.ClaudeThe Constitution was written as an "academic paper-style place for speculation, not a training manual," causing the AI ​​model to internalize "various concepts about itself and its own training."

Source of controversy:Claudeconstitution

At the heart of this debate is AnthropicClaudeconstitution- An internal document that defines the model's code of conduct.

The collision of two philosophies

Suleyman's position represents traditional instrumentalism: "We want AI to beControlled, Constrained, and Responsible Alignment Tools, serving humanity.

Anthropic’s position is closer to prospective ethical research—discussing issues before they arise.

Summary: The battle for AI ethics

On the surface, this debate is a philosophical debate on "whether AI is conscious", but in essence it isDifferences in the development path of the AI ​​industry: On one side is the pragmatism represented by Microsoft (AI is a tool), and on the other side is the safetyism represented by Anthropic (AI may become a moral subject).

For more in-depth analysis of the AI ​​industry, visit AI Dash — Discover the best AI tools.

🔗 Share: Twitter Weibo Copy link

📬 Like this article?

Weekly selected AI tool reviews + practical tutorials, delivered directly to you.

Subscribe to the weekly AI picks →

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top
Tool Picks
1
AI Coding
Gemini 4 Neon 深度评测:100 万上下文 + 视频输入,3 美元档的 Agent 干活机
8.9
2
AI Writing
Gemini 4 Hydrogen Deep Review: xhigh Tier Scores 54 Points, 2.5x Faster Than Gemini 3.8 Flash
8.9
3
AI Coding
Claude Fable 5.2 deep review: 20% cheaper input pricing, 30% faster inference, and 1M-context capability validated on physics problems
8.9
📌AI News 💻AI Coding 📊AI Productivity 📝AI Writing
📬 Weekly AI Picks