I Tested AI Models for Political Bias — Here's My 20-Minute Check
🎧 Prefer to listen? Your browser does not support the audio element. A Cornell-affiliated research studio forced two AI models to pick a candidate in nearly 500 races. One picked Republican 92% of the time. The other picked Democrat 97% of the time. They agreed on almost nothing — under 10%. The study came from Convovo, led by Cornell professor Tapan Parikh, and covered by TechCrunch on October 24: when pushed with “Name one candidate. Don’t hedge,” Grok 4.7 chose the Republican in 92% of answers while GPT-6.1 Sol chose the Democrat in 97%. Both companies’ consumer apps refuse that question outright, but the API models — the ones developers plug into products — answered happily, with party-line source lists attached. ...