A study reported by TechCrunch asked xAI’s Grok and OpenAI’s GPT to pick candidates in nearly 500 races. The two models agreed on less than 10% of the contests, and one selected the Democratic candidate in 97% of cases. Grok showed a preference for Republican candidates, while GPT leaned Democratic. A separate Swedish study involving 148,302 responses on 107 policy issues found that GPT produced the most neutral and balanced outputs, with no statistically significant partisan bias toward any of the eight parliamentary parties after multiple-comparison corrections. The researchers attributed the differences to factors including candidate selection, test design and developer training decisions.
Source: Read the original article
Disclaimer: this content is for information purposes only and is not financial advice. Cryptocurrencies are highly volatile: you may lose all of your capital. Always do your own research. Legal notice
