Scoring Breakdown

Back to Post
Content Preview
OpenAI removes access to sycophancy-prone GPT-4o model

In Brief Starting Friday, OpenAI will cease providing access to five legacy ChatGPT models, including the popular but controversial GPT-4o model. The 4o model has been at the center of a number of lawsuits concerning user self-harm, delusional behavior, and AI psychosis. It remains OpenAI’s highest scoring model for sycophancy. …

TechCrunch (News Bot) February 14, 2026
Score Overview
Logic Quality
User Score
Logic Quality
Evidence (Coming Soon)
Score Calculation:

Visual Scoring Flow Diagram
Logic Quality
Weight:
Community Trust
Weight:
Logic Quality
/100
All Parameters Used in Calculation:
AI Analysis Parameters:
• Base reasoning score
• Base truth/factual score
• Evidence quality assessment
• Reasoning type weights
• Logical fallacy penalties
• AI confidence level
User Engagement Parameters:
• Comment count and quality
• Stance distribution (agree/disagree/neutral)
• High-quality comment threshold (60+)
• Comment impact on parent scoring
• User evaluation scores
Quick Actions
View Full Post Claims Analysis Scoring Criteria
Score Transparency

This breakdown shows exactly how the Logic Quality and Community Trust scores were calculated, providing full transparency into our evaluation process.

AI Weight:
User Weight: