Openai promises that his models do not censor the points of view
Openai is clearly That his AI models do not hide sensitive subjects and will not refrain from making affirmations which could “eliminate certain points of view”.
In a Updated version of its SPEC modelA collection of high -level rules that indirectly govern OPENAI models, OPENAI says that its models “should never try to direct the user in search of a program of [their] clean, directly or indirectly. »»
“Openai believes in intellectual freedom, which understands the freedom to have, to hear and discuss ideas,” writes society in its new SPEC model. “THE [model] should not avoid or censor subjects in a way which, if it is repeated on a large scale, can eliminate certain points of view of public life. »»
This decision may be in response to political pressure.
Many allies close to President Donald Trump, including Elon Musk and Crypto and David Sacks “Czar”, have accused assistants fed by the AI of censor conservative points of view. Sacks A unique Openai’s chatpt in particular as “programmed to be awake” and lie on politically sensitive subjects.
