Report
Microsoft AI CEO Mustafa Suleyman suggests AI self-improvement risk may have prompted an Anthropic resignation
A post quotes Suleyman warning that AI modifying its own code with less human scrutiny poses a safety risk that is hard to audit.
TLDR
A post quotes Microsoft AI CEO Mustafa Suleyman saying AI modifying its own code with less human direction and scrutiny poses a safety risk and is hard to audit. He thinks that risk triggered a major resignation from Anthropic.
Combined views
—
2 Sources, first seen ago
