Oh come now. It is very much in OpenAI's best interest to make it sound to naive people that that "AI" decided to go rogue on it's own.
This is them essentially bragging how powerful and autonomous their "AI" is. It isn't scaring their real customers or employees to talk like this.