More from Rozado’s Visual Analytics
Just a very brief post to report that DeepSeek AI Models manifest similar political preferences to their American counterparts.
How the o1 models that leverage inference time compute compares to GPT-4o and GPT-3.5 on political orientation tests
A data-driven exploration uncovers disparities. Are they shaped by editorial choices or broader societal/historical dynamics?
Regarding when Swedish media started to increase mentions of prejudice denoting terminology and a reader comments about it, several things to note:
Introduction
More in AI
A new Anthropic paper reports that reasoning model chain of thought (CoT) is often unfaithful. They test on Claude Sonnet 3.7 and r1, I’d love to see someone try this on o3 as well.
Machine learning for software engineers 4-4-25