Exploring Why LLMs Show Favoritism Towards Certain AI Companies

Recent analysis using heatmaps reveals that LLMs exhibit favoritism towards their own developers. This bias is likely caused by positive sentiment imbalances in public pre-training data rather than intentional design.

2026-07-20 ~ 2026-07-21 · 2 related posts

• A take on why models may favor some AI companies over others — JacquesThibs · 2026-07-20
• Heatmap results may reflect model bias toward certain AI companies — StephenLCasper · 2026-07-21