The tech overlords want to reshape reality

[Geee]

Spiked

‘This is America, where racism is the No1 value our populace seeks to uphold above all others.’

This is the kind of sentiment you used to only hear in faculty lounges, freshman dormitories and left-wing struggle sessions. But times are changing. These are the words not of some blue-haired student, but of Jack Krawczyk, one of Silicon Valley’s most influential figures.

Krawczyk is the senior director of product management for Gemini, Google’s main group of artificial-intelligence models. Krawczyk, who tweeted that in 2018, was the person primarily responsible for the implosion of the obscenely woke Google Gemini AI last month.

 

Almost immediately after Gemini’s launch, users found that asking it to generate images of historical figures would result in Native American Founding Fathers, South Asian female popes and even black Nazi soldiers. This absurdity was not confined to image generation, either. Gemini’s chatbot also came up with surprisingly charitable descriptions of Adolf Hitler and Mao Zedong, as well as an apologia for paedophilia. The whole episode revealed a moral idiocy on the part of its creators.

The debacle also showed that AI is actually neither artificial nor intelligent. Much of what is called AI is really content generated by a module that has undergone ‘supervised machine learning’. Supervised machine learning is akin to training a dog, in that the outcome is heavily determined by who is doing the supervising. If the dog is trained not to bite anyone, the dog is unlikely to bite. If, on the other hand, you train your dog to bite certain kinds of people, the dog is almost certain to obey its training. In the case of Gemini, Google trained its dog to bite white people – its coding simply would not allow Gemini to produce images that were not racially diverse.