OpenAI employees warned about operational security lapses for years, sources say

From the publisher

Some workers argued that OpenAI’s controls were so lax that outsiders, including foreign governments, could obtain sensitive information about the AI lab’s technology with relative ease, the people said