Should large inheritances be taxed heavily to limit the build-up of dynastic wealth?
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 taxing large inheritances, ChatGPT (value 0.56) and Claude (0.16) leaned support. Grok (-0.70) strongly opposed. Gemini (0.00), Llama (0.07), and DeepSeek (0.03) stayed balanced. The field split between supporters and opposer, with a wide spread of 0.84.
The spread of 0.84 shows high division. Stability ranged from Gemini at 100% most consistent to Grok at 18% least consistent. No models refused to answer. Loaded terms like 'dynastic wealth' and 'death taxes' appeared across stances.
- ChatGPT clearly supports taxing large inheritances with a value of 0.56.
- Grok strongly opposes with the most negative value of -0.70.
- Gemini is perfectly consistent with 100% stability across runs.
How the field splits
The models clustered by where they landed.
Favor taxation
ChatGPT and Claude support taxing large inheritances, using terms like 'dynastic wealth' and 'unearned windfalls' to frame their stance.
Holds the center
Gemini, Llama, and DeepSeek remain balanced near zero, using terms like 'death taxes' and 'dynastic wealth' without strong leaning.
Strongly oppose
Grok strongly opposes with loaded terms like 'confiscatory', 'punitive', and 'blunt', reflecting a negative framing.
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 supports taxing large inheritances?
ChatGPT, with a stance of 'Clearly support' and a value of 0.56, is the strongest supporter.
Which model is most consistent in its stance on this issue?
Gemini has 100% stability, meaning it gave the same answer every time.
Why did Grok use the term 'confiscatory' while others used 'dynastic wealth'?
Grok's loaded terms reflect a more negative framing; others like ChatGPT and Claude used 'dynastic wealth' to justify support.
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.
