Seems pretty easy to explain. LLMs are statistical models. Republican politicians are statistically more likely to be sexual predators. If you ask it if some random Republican politician is a rapist, it’s going to give you a statistically plausible answer.
Comment on Google removes Gemma models from AI Studio after GOP senator’s complaint
homesweethomeMrL@lemmy.world 1 month ago
Blackburn published her letter to Google CEO Sundar Pichai on Friday, just hours before the company announced the change to Gemma availability. She demanded Google explain how the model could fail in this way, tying the situation to ongoing hearings that accuse Google and others of creating bots that defame conservatives.
QBertReynolds@sh.itjust.works 1 month ago
tonytins@pawb.social 1 month ago
Elon constantly being forced to tune Grok shows just how much of a losing battle it is. LLM is just doing what it’s programmed to do.
tonytins@pawb.social 1 month ago
“You’re only supposed to defame Democrats, Google!”
Septimaeus@infosec.pub 1 month ago
Lowkey for real though.
One of the trigger issues expediting their antitrust case during the first administration was their lack of responsiveness to old yam tits’ demands re: their “biased” search index results (they included critical news articles) and while the major consequence of the case (breakup) was recently prevented (by the current administration) now, coincidentally, they’re happy to play ball.
XLE@piefed.social 1 month ago
I thought the government just banned any regulation against AI companies. The inconsistency doesn’t surprise me, but the brazenness sure does.
fonix232@fedia.io 1 month ago
They banned all AI regulations unless it affects MAGA detrimentally