Sources: OpenAI repeatedly dismissed internal warnings about inadequate monitoring of testing, prioritizing fast releases without additional security protocols (New York Times)

Sources indicate that OpenAI repeatedly disregarded internal warnings regarding insufficient monitoring of testing. The company reportedly prioritized rapid releases over implementing additional security protocols. Both employees and security researchers had cautioned OpenAI about the safe testing of its AI models and the need to strengthen security measures.