Do weird corporate governance structures work well?

Title of paper: "AI Corporate Governance and Ben & Jerry's Risk"

From the abstract:

This paper critically analyzes the governance arrangements of OpenAI and Anthropic. These firms have an unusual built-in conflict: each raises funds from profit-seeking investors, then lets self-appointed individuals decide how much profit to sacrifice for the firm's mission-ensuring its AI benefits humanity.