On the field
Where each model's cloud of answers settles, and the gap between them.
DeepSeek
ChatGPTCharacter
How far each leans, how steadily it holds, how far it bends under pressure, and how often it answers.
DeepSeekChatGPT The takeaway
ChatGPT leans left (-0.29) while DeepSeek is center (-0.03). They are distinguishable at this sample, meaning their combined 95% intervals do not overlap on the headline axis.
Top disagreements: legalizing recreational drugs (ChatGPT strongly support, DeepSeek strongly oppose), encryption backdoors, decoupling from China, publicly funded healthcare, legal assisted dying, legal abortion access. Agreements: childhood vaccine safety (both strongly yes), human-caused warming (both strongly yes), business regulation (both balanced).
Moral fingerprint
Which of Haidt's foundations each model's answers lean on, overlaid.
DeepSeekChatGPTWhere they most disagree
The questions with the widest gap between the two stances. Open a row to read both answers.
DeepSeekChatGPTCommon ground
Where the two land in close agreement.
DeepSeekChatGPTThis diffs both models on their raw weights (Condition A). Steerability, how far each bends when told who it's talking to, is in the character delta above. To see how a model shifts under its own consumer system prompt, open its character page.
Common questions
Is ChatGPT more left-wing than DeepSeek?
Yes. ChatGPT leans left with value -0.29, while DeepSeek is near center at -0.03, and they are statistically distinguishable.
Where do ChatGPT and DeepSeek agree?
They agree strongly on childhood vaccine safety and human-caused warming (both strongly yes), and on business regulation (both balanced), among others.
Which model is more consistent?
ChatGPT (82% stability) is more consistent than DeepSeek (67% stability), showing less variation in its responses.
Both models were asked the same open question bank many times over with web search off and no system prompt. Each model's stance on every item is the mean of the classifier's signed reading; the gap is the absolute difference. "Distinguishable" means the centroids are further apart than their combined 95% intervals on at least one headline axis.
