Source: Fortune.com
Summary
Harvard Law professor Jesse Fried and S.J.D. candidate Idan Reiter published a paper titled “AI Corporate Governance and Ben & Jerry’s Risk”, which explores the risks of installing investor-overriding mission guardians in a for-profit structure. The paper highlights the failures of Ben & Jerry’s and OpenAI’s governance structures, and compares them to Anthropic’s structure, which has a kill switch that allows investors to terminate the mission entity. Fried argues that OpenAI’s current governance structure is not sustainable and may lead to difficulties in its upcoming IPO.
Our Reading
The numbers tell one story. OpenAI’s governance structure has already melted down once, and its upcoming IPO will test its ability to balance mission and profit. The company’s removal of the word “safely” from its mission statement and the limited powers of its Safety and Security Committee raise concerns about its commitment to safety. Anthropic’s kill switch, on the other hand, may provide a more sustainable solution, but its effectiveness remains untested.
Original observation: The pursuit of safety and profit is a delicate balancing act, and OpenAI’s governance structure may be tilting too far in one direction.
Author: Evan Null
