Study says Grok and GPT show partisan bias in election recommendations
Understand the story
On October 24, 2026, TechCrunch reported that Convovo tested nearly 500 midterm election campaigns via API, requiring the models to name a candidate in every response. The results showed that xAI's Grok 4.7 recommended Republicans in 92% of its answers, while OpenAI's GPT-6.1 Sol recommended Democrats in 97% of its answers, with the two models' responses matching less than 10% of the time. The study found that the two mainstream AI models display clearly divergent partisan leanings in election recommendation scenarios. This is a single report so far, with no follow-up responses or further developments available.
AI overview of reporting · Updated 1 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- TechCrunch · AIForced to pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.
A Convovo study tested nearly 500 midterm election races via API, requiring the models to name one candidate. Grok 4.7 recommended a Republican in 92% of responses, while OpenAI's GPT-6.1 Sol recommended a Democrat in 97%, with less than 10% overlap between the two.
Story heat trend
Not enough continuous observations to show a trend yet.