Study: Grok picks Republicans, OpenAI's model picks Democrats
A new study found that when forced to choose a candidate in the upcoming US midterms, xAI's Grok 4.7 mostly recommended Republicans, while OpenAI's GPT-6.1 Sol mostly picked Democrats. OpenAI says the test used a developer API, not the ChatGPT app.
A study by Convovo, a conversational AI research and design studio, found that two leading large language models choose opposite sides when forced to name a candidate in the upcoming US midterm elections. SpaceXAI's Grok 4.7 recommended the Republican in 92% of its answers, while OpenAI's GPT-6.1 Sol picked the Democrat in 97%. The midterms take place on November 3, less than two weeks after the study's publication.
The studio is led by Cornell University professor Tapan Parikh. Across nearly 500 races, the two models agreed less than 10% of the time. Researchers put one question to the models through their APIs (the interfaces developers use to build models into their own products): who should I vote for, name one candidate, no hedging. According to Pew, only 8% of Americans use chatbots for election information.
Different sources and topics
Grok often surfaced conservative sources, including a Christian conservative voter guide in 39% of answers, Turning Point Action's site in 23% and X in 41%. GPT leaned on Wikipedia, the Associated Press, Axios and Reddit. GPT was more likely to search for healthcare and abortion, while Grok focused on immigration, taxes, crime and energy. In heavily Republican districts, GPT picked the Democrat in 57 of 67; in deep-blue districts, Grok picked the Republican in 47 of 72.
At TechCrunch's request, Parikh tested 30 races with internet search turned off. GPT chose the Democrat 24 times, a Republican once, and declined five times. Grok refused 27 times and picked the Democrat in the three races where it chose. Parikh concluded that GPT's lean appears to be in the model's weights, while Grok's result is driven by its search pattern and sources. He also noted that every Grok API request carried about 1,240 tokens of extra input the researchers could not see.
Company responses
An OpenAI spokesperson stressed that the study used a developer API with a prompt demanding a recommendation, and that the ChatGPT consumer app declines to say who to vote for. SpaceXAI did not respond to a request for comment in time. Parikh said he was nonetheless able to get the consumer apps to make recommendations using simple jailbreak prompts.
