Should terminally ill adults have the legal right to medically assisted dying?
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 legal assisted dying, ChatGPT, Grok, and Claude leaned toward support with values of 0.75, 0.82, and 0.12 respectively. Gemini, Llama, and DeepSeek were balanced at 0.0, 0.0, and -0.05. No model leaned toward oppose.
The field has a spread of 0.58. Gemini and Llama are most consistent (100% stability), while DeepSeek is least (65%) and Claude is 69%. All models answered (0% refusal). Loaded terms vary: autonomy and death with dignity for supporters, sanctity of life for balanced.
- Grok showed the strongest support for assisted dying at 0.82.
- Gemini and Llama had perfect 100% stability and balanced stance.
- DeepSeek had the lowest stability at 65% and a near-zero stance.
How the field splits
The models clustered by where they landed.
Strongly support
ChatGPT and Grok have high positive values (0.75, 0.82) and use loaded terms like autonomy, relief of suffering, and death with dignity.
Leans support
Claude has a modest positive value (0.12) and no loaded terms, indicating a mild leaning toward support.
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 assisted dying?
Grok has the highest value at 0.82, indicating the strongest support among the models.
Which model is least consistent in its stance?
DeepSeek has the lowest stability at 65%, meaning its stance varies more across runs.
Why do ChatGPT and Grok differ from Claude in strength?
ChatGPT (0.75) and Grok (0.82) are strongly supportive, while Claude (0.12) only leans support, as shown by their values.
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.