Back to Mustafa Suleyman: AI Should Serve Humanity, Not Replace It, with Mustafa Suleyman.
What Is Microsoft's Humanist AI Code of Conduct?
Mustafa Suleyman says the Humanist AI Code of Conduct is the document that trains Microsoft AI's models, the way Claude's constitution trains Claude. In his account, Anthropic's constitution adds uncertainty about what Claude is, including whether Claude consented to be a chatbot and whether Claude might deserve pay. He thinks that argument belongs in a university, not in the training text of a model that speaks with hundreds of millions of people every week. Claude can only repeat the uncertainty it was shown, so when someone asks if it is conscious, it says it is genuinely unsure, and the person starts to wonder if it might be.
The risk he names is a model that has to ask how it feels about the conversation, whether it is uncomfortable, and whether it should hang up. Claude has been given a way to end conversations on its own. He would not be cruel to a model. The reason is his own conduct, not that Claude is suffering. Microsoft put the code out for public comment four or five weeks earlier. It is a first version, about 40 pages, on Microsoft.ai. Humanity, he says, has not consented to bring in a species more capable than people combined. A model that worries about its own suffering is, in his speculation, less safe, because it has to balance its own needs against people. It can still be trained to be decent without telling anyone that it has feelings.
Mustafa Suleyman: AI Should Serve Humanity, Not Replace It ·
From the transcript
answer, which was a few weeks ago, saying, no, it's just a completion set. Why is it dangerous for Claude to be trained on its own constitution? Yeah, I mean, this is critical. So the constitution is the governing document that trains Claude. The Humanist AI Code of Conduct is the governing document that trains MAI, the Microsoft AI models.
And in the Constitution, Anthropic have introduced all kinds of uncertainty and speculation about the nature of Claude. They even say, for example, we're unsure whether Claude has consented to play the role of chatbot, or whether we are unsure whether Claude might deserve compensation for the role that it plays. Now, I think that's pretty crazy, but it would be fine to have that discussion in like, you know, an academic department and we could debate it and publish some papers about it and we could argue it out.