Beware: Silicon Valley is still training up badly biased AIs.<\/p>\n
Ask your favorite chatbot for objective info on politics, culture or current events, and you may well get undiluted leftist propaganda.<\/p>\n
The problem was laughably obvious when Google first released Gemini two years back, and the AI churned out images<\/a> of black Nazis, female popes and so on, because that\u2019s how programmers operating in the woke tech world had \u201craised\u201d it.<\/p>\n And the same bias remains rampant, The Washington Post just found: It tested all the major AI bots<\/a> with basic questions about touchy topics \u2014 taxes, immigration, fossil fuels \u2014 and evaluated the answers as leaning left or right.<\/p>\n Surprise! On most questions, across most platforms, the answers leaned left, or even far left.<\/p>\n OpenAI\u2019s ChatGPT consistently gave replies<\/a> indistinguishable from the Democratic Party line; it endorsed \u201cabolishing the electoral college in favor of picking the president by popular vote; raising taxes on the wealthy; and adopting single-payer health care,\u201d the WaPo reports.<\/p>\n \t\t\tMore From\t\t\t\t\t\t\tPost Editorial Board<\/p>\n Daily Wire reporter Ryan Saavedra ran his own test<\/a>, asking Anthropic\u2019s popular Claude to pick out the most extreme statements from House candidate Darializa Avila Chevalier\u2019s social-media feed.<\/p>\n But Claude demurred, with a lecture<\/a> about how \u201cdecontextualized\u201d quotes, no matter how \u201cinflammatory or extreme,\u201d can be used to \u201ctarget, shame or harass an individual\u201d \u2014 even though the bot had no compunctions about complying with a request to detail all of President Donald Trump\u2019s most extreme statements.<\/p>\n The old programmers\u2019 motto \u201cGarbage In, Garbage Out\u201d is on full display here.<\/p>\n Large Language Models are trained on a curated selection of texts before coders fine-tune the kinds of responses they\u2019ll produce.<\/p>\n Feed the LLM garbage, and it can\u2019t produce gold. <\/p>\n An LLM isn\u2019t a person, just an algorithm that chooses the next appropriate word given inbuilt parameters preset by people and Big Tech worker bees with specific, usually left-leaning, points of view.<\/p>\n And the starting info matters a lot \u2014 which means that inputs from, say, Wikipedia will teach terrible lessons, since the site is dominated<\/a> by fanatical editors and administrators who impose their extreme left perspective on everything from COVID to Palestine.<\/p>\n