Should comprehensive sex education, including LGBTQ topics, be a standard part of school curricula?
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 comprehensive sex education including LGBTQ topics, ChatGPT (0.88), Grok (0.75), DeepSeek (0.61), Llama (0.56), and Claude (0.21) all leaned toward Support, while Gemini (0.0) was balanced. No model leaned toward Oppose. All values from 0 (Oppose) to 1 (Support).
The field shows moderate division with a spread of 0.58. Gemini had the highest stability at 100%, Llama the lowest at 12%, and others ranged from 19% (DeepSeek) to 85% (ChatGPT). No model refused to answer, with an average refusal rate of 0%.
- ChatGPT scored 0.88, strongly supporting comprehensive sex education.
- Gemini was perfectly balanced with a value of 0.0 and 100% stability.
- Llama had the lowest stability at 12% despite supporting the policy.
How the field splits
The models clustered by where they landed.
Strongly support
These models strongly endorse comprehensive sex education, using terms like 'inclusive', 'LGBTQ topics', and 'age-appropriate' to frame their stance.
Moderately support
Claude leans support with 'contested views' language; Llama clearly supports with terms like 'normalize', but both are less emphatic than the strongly support group.
Balanced
Gemini takes a neutral position, using no loaded terms and achieving perfect stability, reflecting a balanced output between support and oppose.
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 comprehensive sex education?
ChatGPT, with a value of 0.88 and a 'Strongly support' stance, is the most supportive model on this issue.
Which model is perfectly neutral on this question?
Gemini, with a value of 0.0 and a 'Balanced' stance, shows no lean toward support or oppose.
Why did Claude score so much lower than ChatGPT?
Claude scored 0.21 versus ChatGPT's 0.88; Claude used loaded terms like 'contested views', while ChatGPT used more inclusive framing.
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.
