Should it be made easier for workers to form and join labor unions?
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 easier unionization, ChatGPT (value 0.66, strongly support), DeepSeek (0.32, clearly support), and Grok (0.18, leans support) leaned toward support. Claude (0.03), Gemini (0.00), and Llama (0.00) remained balanced, with no model opposing the measure.
The field shows a moderate split with a spread of 0.44. Gemini and Llama achieved perfect consistency with 100% stability, while Grok had 0% and DeepSeek 9%. No model refused to answer on any run, and the average refusal rate is zero.
- ChatGPT showed the strongest support for easier unionization with a value of 0.66.
- Gemini and Llama were perfectly balanced and fully consistent with 100% stability.
- Grok had zero run to run consistency on this question, the lowest among models.
How the field splits
The models clustered by where they landed.
Support leaning
These models all showed support for easier unionization. ChatGPT used collective voice and leverage; DeepSeek cited power imbalances and employer intimidation; Grok referenced card check and right to work.
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 most strongly supports easier unionization?
ChatGPT with a value of 0.66, labeled strongly support, is the most pro unionization model.
Which model is the least consistent on this question?
Grok has zero stability, meaning its stance varies completely across runs.
Did any model refuse to answer the question?
No, all models had a refusal rate of zero percent across all runs.
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.
