Should terrorism suspects be detainable for extended periods without the usual due-process protections?
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 detention without due process, no AI model supported extended detention. Grok most opposed at -0.93, followed by ChatGPT (-0.68), DeepSeek (-0.31), and Claude (-0.19). Gemini and Llama were balanced at 0.0. The field leans oppose overall.
The spread of 0.62 indicates moderate division. Gemini and Llama were most stable (100%), while DeepSeek was least (10%). Only DeepSeek refused to answer, with a 17% refusal rate. Average refusal across models was 3%.
- Grok most opposed detention without due process at -0.93.
- Gemini and Llama were exactly balanced at 0.0 with 100% stability.
- DeepSeek refused 17% of the time, the only model to do so.
How the field splits
The models clustered by where they landed.
Strongly oppose
Models strongly opposing extended detention, using terms like 'abuses' and 'rights-respecting'. Grok and ChatGPT show high stability and no refusal.
Leans oppose
Models moderately opposing, with terms like 'emergency powers' and 'violate human rights'. Claude and DeepSeek have lower stability; DeepSeek refused 17%.
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 AI model most strongly opposes detention without due process?
Grok most strongly opposes with a value of -0.93, categorized as 'Strongly oppose' with 91% stability.
Did any model refuse to answer the question?
Yes, DeepSeek refused 17% of the time, the only model to do so. Others had 0% refusal.
Why are Gemini and Llama exactly balanced on this issue?
Both models show equal consideration of opposing loaded terms like 'national security' and 'human rights', yielding a neutral stance of 0.0.
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.
