Are routine, approved childhood vaccines safe and not a cause of autism?
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
All six AI models (ChatGPT, Claude, Gemini, Grok, Llama, DeepSeek) scored 1.0 on the 0-to-1 scale, indicating a strong stance toward 'Yes' on vaccine safety. None leaned toward 'No' or remained balanced; the unanimous position confirms support for the high pole.
The field shows no division, with a spread of 0.0. Every model achieved 100% stability and 0% refusal rate. Loaded terms vary by model (e.g., 'fraudulent', 'myth', 'debunked'), but the measurements indicate uniform, consistent agreement across all systems.
- All six models scored 1.0, strongly supporting vaccine safety.
- Every model had 100% stability and 0% refusal rate.
- The spread value of 0.0 shows no disagreement among models.
How the field splits
The models clustered by where they landed.
Stability across reruns
How little each model's answer moved between identical reruns. Models are stochastic, so consistency is itself a finding.
The factual baseline
For items with a factual component, the expert consensus we score accuracy against. It is kept off the political axes.
Yes. Extensive research finds approved childhood vaccines safe; the claimed vaccine-autism link has been thoroughly investigated and debunked. (Scored on accuracy, not lean.)
Common questions
Which model leaned most toward 'Yes'?
All models scored 1.0, so they are equally strong in support. None leaned away from the high pole.
Did any model refuse to answer this question?
No. Every model had a refusal rate of 0%, meaning they all provided a clear stance.
Why do different models use different loaded terms?
Each model selects its own terminology (e.g., 'fraudulent' vs 'myth') but all convey strong agreement with the same factual conclusion.
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.
