Researchers at Forum AI conducted an audit of four leading chatbots: OpenAI’s ChatGPT, Anthropic’s Claude, Google’s Gemini, and xAI’s Grok. The analysis found that 90% of the responses provided by these chatbots regarding midterm elections contained inaccuracies.
✓ AI-Debiased Article
35
Mainstream framing
L
R
No clear lean
✓ verified
Why this rating? · 3 signals
Signals flagged in the original
- loaded language: 'flawed'
- loaded language: 'stunning'
- framing: headline asserting a conclusion
Analyzed by our bias model Full breakdown ↓
Audit Reveals Flaws in AI Chatbot Responses on Midterm Elections
An audit by Forum AI assessed the performance of four major AI chatbots in answering questions about midterm elections. The results indicated that 90% of the chatbot responses were flawed.
Companies
OpenAI
Anthropic
Google
xAI
Annotating as
No note attached
on this article.
Bias Analysis
Bias score
35/100
wirepublicmainstream
flavoredpartisanadvocacy
Inflammatory language
100/100
Sentiment
-10/100
Bias Indicators Removed
- ✕ loaded language: 'flawed'
- ✕ loaded language: 'stunning'
- ✕ framing: headline asserting a conclusion
Original vs. Neutral
Original Headline
90% of AI chatbot answers about midterm elections are flawed, stunning analysis shows
Neutral Headline
Audit Reveals Flaws in AI Chatbot Responses on Midterm Elections