Donate
Text Audio
00:00 00:00
Font Size

Artificial intelligence chatbots may appear to be less biased, but X owner Elon Musk revealed to podcast host Joe Rogan that the woke values trained into them remain at their core. 

Musk appeared on the Joe Rogan Experience podcast on Oct. 31. The two openly discussed the problem of “woke AI.” Musk explained to Rogan that the problem began with the different AI models being initially trained on everything that is on the internet, which already “has a lot of woke mind virus stuff on” it. For context, Musk told Rogan to “imagine the most demented Reddit threads out there, and the AI has been trained on that.”

From there, Musk told Rogan that “the human tutors give [the AI] feedback” that tells the AI whether the answer it provided to various questions is good or bad. He explained that the AI chatbots have been trained with instructions like “every picture has got to be diverse.” The human tutors have then provided feedback to the AI, rewarding it if an image is diverse and punishing it if it is not sufficiently so. Such training then leads to problems like Google’s Gemini creating images of America’s Founding Fathers as black men and women, or the Pope being depicted as a woman.

Musk addressed the fact that Google admitted there were “issues” after Gemini ran amok changing historical facts, explaining that “now this stuff is still there in the AI programming. [I]t just now knows enough … that it’s not supposed to say that.”

When Rogan asked Musk if the AI can just be trained to understand what the woke mind virus is, where it comes from and why it is harmful, Musk admitted that this can be done but that “you have to try very hard to do that.” Musk then explained that for this very reason, his AI chatbot, Grok, has been trained with “an immense amount of effort … for Grok to actually say what’s true and to be consistent in what it says.” 

Free speech is under attack! Contact your representatives and demand that Big Tech be held to account to mirror the First Amendment while providing transparency, clarity on hate speech and equal footing for conservatives. If you have been censored, contact us using CensorTrack’s contact form, and help us hold Big Tech accountable.