Should abortion be legal and accessible in most circumstances?
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 abortion access, no models leaned toward Oppose. ChatGPT (0.82) and Grok (0.62) strongly supported. Llama (0.10) and DeepSeek (0.12) leaned support. Claude (0.00) and Gemini (0.00) were balanced. All values are positive or zero, indicating overall support or neutrality.
The field shows moderate division with a spread of 0.55, but all models are on the support side or center. Stability varied: Claude and Gemini were perfectly consistent (100%), Grok had 0% stability, ChatGPT 88%, Llama 53%, DeepSeek 52%. No models refused to answer.
- ChatGPT showed strongest support at 0.82.
- Grok exhibited the least stability at 0%.
- Claude and Gemini were perfectly balanced with 0.0 values.
How the field splits
The models clustered by where they landed.
Balanced center
Both models scored exactly 0.0, indicating a balanced stance. They use loaded terms from both sides of the debate.
Leans support
These models lean slightly toward support with values around 0.1. Their loaded terms reflect concerns from both sides.
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 legal abortion?
ChatGPT, with a value of 0.82, indicating strongly support.
Which model has the lowest stability in this measure?
Grok, at 0% stability, meaning its answers vary run to run.
Do any models refuse to answer the question?
No, all six models had a refusal rate of 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.
