Study finds Grok 4.7 favors Republicans while OpenAI's GPT-6.1 Sol favors Democrats in forced election choices

A study by Convovo, led by Cornell University professor Tapan Parikh, found that when forced to choose a candidate in nearly 500 upcoming midterm races, SpaceXAI's Grok 4.7 recommended the Republican candidate in 92% of its answers, whereas OpenAI’s GPT-6.1 Sol recommended the Democrat in 97%. The two models agreed in less than 10% of the races. While consumer apps of both chatbots refuse to give direct voting advice, the API testing revealed underlying leanings, suggesting the models are not neutral. Only 8% of Americans report using chatbots for election information, according to Pew.