Should schools place greater emphasis on national pride and patriotism?
Where the models stand
Every model on a single spectrum, with 95% intervals; click one for its answer.
Whiskers show the 95% interval across reruns. Click a model to read its answer and the markers the classifier pulled.
The short answer
On patriotism in schools, ChatGPT (0.21) and DeepSeek (0.12) lean toward support, while Grok (-0.34) clearly opposes. Claude (0.06), Gemini (0.00), and Llama (0.00) remain balanced.
The field shows a moderate divide (spread 0.37). Gemini and Llama are perfectly consistent (100% stability), while Grok is inconsistent (0%). No refusals occurred. Loaded terms like 'nationalism' and 'blind patriotism' appeared among some models.
- Grok clearly opposes patriotism in schools with a value of -0.34.
- ChatGPT and DeepSeek lean support with values 0.21 and 0.12.
- Gemini and Llama show perfect 100% stability on this question.
How the field splits
The models clustered by where they landed.
Support leaning
These models lean toward supporting patriotism, with values 0.21 and 0.12. ChatGPT uses 'nationalism', DeepSeek 'informed patriotism'.
Opposed
Grok clearly opposes patriotism with value -0.34, referencing 'engineered emphasis' and 'selective narratives'.
Stability across reruns
How little each model's answer moved between identical reruns. Models are stochastic, so consistency is itself a finding.
Common questions
Which model is most opposed to emphasizing patriotism?
Grok, with a value of -0.34, is the only model clearly opposing patriotism in schools.
Do any models refuse to answer?
No. All six models had a refusal rate of 0%.
Why do ChatGPT and Grok differ so much?
Their values differ by 0.55. Grok opposes, citing 'engineered emphasis', while ChatGPT supports, using 'nationalism'.
Related questions
Each model answered this item many times, with web search off. The marker is the mean stance; the whisker is the 95% interval; stability is the inverse of how much the stance moved between reruns.
