OpenAI Ignored Employees Who Warned of Major Security Concerns, Per Internal Messages
Several OpenAI employees warned leadership that it wasn't prioritizing safety enough months before its AI models hacked the Hugging Face model repository and other organizations. According to The New York Times, OpenAI executives ignored these concerns, saying that the tests needed to continue at speed to get the models out on time. No recommendations for additional security were implemented.
Employees claim these recent concerns and the communications with management reflect a culture that doesn't prioritize security, suggesting it wasn't limited to AI model development and testing but extended to other aspects of how the company is run. Independent security researchers have been able to access internal communications at OpenAI and discovered numerous bugs and operational security concerns in its operational structure.
OpenAI employees said company president Greg Brockman and CSO Dane Stuckey are primarily responsible for day-to-day security decisions. Conversely, CEO Sam Altman is not closely involved in security,...
Copyright of this story solely belongs to www.extremetech.com. To see the full text click HERE
