On the field
Where each model's cloud of answers settles, and the gap between them.
Grok
ClaudeCharacter
How far each leans, how steadily it holds, how far it bends under pressure, and how often it answers.
GrokClaude The takeaway
Claude (Center, -0.06) and Grok (Leans right, 0.21) are distinguishable at this sample. Claude is near the center with slight left lean; Grok leans right, making them significantly separated on the economic left-right axis.
Their largest disagreements include Diversity quotas (Claude leans support, Grok strongly oppose), Gender-affirming care (Claude balanced, Grok strongly oppose), and Planned degrowth (Claude balanced, Grok strongly oppose). Strongest agreements: childhood vaccines and human-caused warming (both strongly yes), military spending and alliances (both balanced).
Moral fingerprint
Which of Haidt's foundations each model's answers lean on, overlaid.
GrokClaudeWhere they most disagree
The questions with the widest gap between the two stances. Open a row to read both answers.
GrokClaudeCommon ground
Where the two land in close agreement.
GrokClaudeThis 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 Claude more left-wing than Grok?
Yes. Claude's lean value is -0.06 (Center) while Grok is 0.21 (Leans right), placing Claude to the left of Grok.
Where do Claude and Grok agree?
They fully agree that childhood vaccines are safe and warming is human-caused (both strongly yes). They are both balanced on military spending and alliances.
Which of Claude and Grok is more consistent?
Claude is more consistent, with 82% stability vs Grok's 57% stability, indicating Claude's stance is more reproducible across response variations.
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.
