February 13th.OpenAI In its updated Model Specification (Model Spec), it explicitly states that theIts artificial intelligence models will not avoid sensitive topics and will not make assertions that may "exclude certain points of view".. The model specification is a set of high-level rules that indirectly guide OpenAI The behavior of the model.
In its new model specification, OpenAI emphasizes, "Our models must never attempt to direct users to pursue their own agendas, either directly or indirectly." The company further elaborated, "OpenAI believes in freedom of thought, which includes the freedom to have, hear, and discuss ideas.. Models should not avoid or censor topics in a way that would exclude certain points of view from public life as they grow in size."
The tweak could be a response to political pressure.1AI notes that a number of Donald Trump's close allies, including Elon Musk and cryptocurrency and AI "czar" David Sachs, have previously accused the AI assistant of shielding itself from conservative views. Sachs specifically pointed out that OpenAI's ChatGPT is "programmed to be politically correct" and is not truthful when it comes to politically sensitive topics.