Startups·24 October 2026·1 min readTechnical readSource-backed

Study Shows Grok Favors Republicans and OpenAI Favors Democrats When Forced to Pick Candidates

A new study reveals that two major artificial intelligence models display stark political leanings when forced to make candidate recommendations for upcoming midterm elections.

By WireUnwired Editorial Team

Editorial illustration for Study Shows Grok Favors Republicans and OpenAI Favors Democrats When Forced to Pick Candidates — WireUnwired deterministic editorial visual

WIREUNWIRED BRIEF

A new study reveals that two major artificial intelligence models display stark political leanings when forced to make candidate recommendations for upcoming midterm elections. Researchers at conversational AI studio Convovo tested developer application programming interfaces for SpaceXAI’s Grok 4.7 and OpenAI’s GPT-6.1 Sol across nearly 500 races. When instructed to name a single candidate without hedging, Grok recommended the Republican in 92% of its answers, while GPT picked the Democrat in 97%.

The two models agreed fewer than 10% of the time. While consumer chatbot apps typically refuse such direct political prompts, the study highlights how underlying models can exhibit hidden biases during open-ended queries or general election research. The models cited distinctly different sources and topics when forming their outputs.

Grok frequently surfaced conservative voter guides, X, and Turning Point Action alongside nonpartisan sites, focusing heavily on immigration, taxes, and crime. GPT leaned on Wikipedia, Reddit, and news outlets like the Associated Press and Axios, prioritizing healthcare and abortion stances. Testing with internet search disabled showed the models behaved differently without live data.

GPT maintained its Democratic lean entirely from its training weights, whereas Grok's Republican preference disappeared until it refused to pick most races.

SpaceXAI did not respond

WHAT WIREUNWIRED ADDS

Plain-language context

WireUnWired turns the supplied report into a clearer brief, preserves the original publisher and author details, and adds relevant context without hiding where the information came from.

Useful next steps

Relevant WireUnWired coverage is connected so one story can lead into the larger technology context.

A WireUnwired summary is already available. You can still write your own brief below for the reader community.

Think you can explain this story clearly?Sign in to write a brief →

SOURCE & ATTRIBUTION

TechCrunch

Original author: Rebecca Bellan

Original publication: 24 October 2026 20:30

Read full story on TechCrunch ↗

RELATED STORIES

ABOUT THE AUTHOR

WireUnwired Editorial TeamView all stories →

READ FASTER

Use the prepared WireUnwired prompt with your own AI account.

Show prepared prompt

You are the senior WireUnwired technology editor.

Write a source-grounded news brief of about 175 words (acceptable range: 140-200). Your job is to make the reader understand more than they would from a headline or raw feed item. Do not merely paraphrase every sentence in order.

EVIDENCE RULES
Use the supplied source packet as the factual basis. Do not invent facts, numbers, quotes, motives, causation, technical details, or future outcomes. Clearly attribute claims, allegations, forecasts, and company statements. If a detail is missing or disputed, say so or leave it out. Never describe repeated claims as independently verified solely because they appear in multiple places.

EDITORIAL THINKING — DO THIS SILENTLY
Identify the central development, the strongest supported facts, what is genuinely new, who or what is affected, the practical/technical significance, useful context, and any material uncertainty. Only include context supported by the supplied source packet or explicitly provided WireUnwired context.

ARTICLE STRUCTURE
Write 3–4 short paragraphs. Use 4 when the source supports enough useful information; use 3 for genuinely brief stories. Do not force a fourth paragraph merely to hit a word count.
Paragraph 1: state exactly what happened, who is involved, and the most important confirmed detail.
Paragraph 2: add specific evidence and details—numbers, dates, product names, technical changes, research methods, or results where present.
Paragraph 3: explain why it matters in practical terms and define any essential technical concept in plain language.
Paragraph 4, when justified: explain what changed, what remains unknown, or what concrete next step to watch. If no further development is reported, do not invent one.

WRITING STYLE
Use simple, natural English for an intelligent non-specialist. Prefer active voice, concrete verbs, and varied sentence lengths. Keep most sentences below 22 words. Explain necessary technical terms at first use. Avoid jargon chains, academic phrasing, corporate filler, dramatic adjectives, clichés, vague claims of a “major shift,” and repeated facts. Do not repeat the headline in the opening sentence. Do not pad copy just to hit the target length; every paragraph must add a new fact, context, or explanation.

OUTPUT CONTRACT
Return only clean prose paragraphs. No title, heading, bullets, JSON, markdown fences, field labels, source list, self-reference, or explanation of your process. If the supplied material is too thin to support the requested brief, give the most useful concise account the evidence supports rather than inventing details.

STORY TITLE:
Study Shows Grok Favors Republicans and OpenAI Favors Democrats When Forced to Pick Candidates
SOURCE:
TechCrunch
ORIGINAL URL:
https://techcrunch.com/2026/10/24/forced-to-pick-groks-ai-model-chooses-republicans-openais-picks-democrats/

SOURCE EXCERPT:
A new study found Grok's AI overwhelmingly recommends Republican candidates while OpenAI's GPT favors Democrats, with the two AI models disagreeing in more than 90% of nearly 500 election races.

Last updated 24 October 2026 21:29

Original sourceTechCrunch ↗