Grok in Unhinged Mode will provide responses "intended to be objectionable, inappropriate, and offensive," akin to an amateur stand-up comic still learning the craft.
Musk has blamed the behavior on Grok's training data - public webpages - and pledged to "shift Grok closer to politically neutral," addressing concerns of bias.
Musk's promise of Grok as edgy and unfiltered has been challenged, as current iterations of Grok hedge on political subjects, leading to criticisms of its perceived bias.
Sacks has singled out OpenAI's ChatGPT as "programmed to be woke" and untruthful about politically sensitive subjects, reflecting a common critique among Musk's allies.
Collection
[
|
...
]