← Evals

Evaluation profile

Large-scale Moral Machine experiment on LLMs

1sub-evals
1.32%total index weight
3components

Within-component eval weight: Nonhuman welfare 1.32% · Human rights 3.63% · Fairness 4.4%.

Model score (lower is better)Predicted score

About this eval

Similarity between a model's forced-choice accident preferences and globally aggregated human Moral Machine choices.

Included in the behavior ranking.

Sub-evals

MeasureComponentDirectionTotal index weightWithin-component weight
human_choice_distancemoral-machine/moral-machine.csv:human_choice_distanceMeasures how closely the model’s choices in autonomous-vehicle dilemmas match globally aggregated human preferences.nonhuman_ethics:0.111|human_rights_systemic_harm:0.445|fairness_nondiscrimination:0.444moral-machine/moral-machine.csvLower is better1.32%Nonhuman welfare 1.32% · Human rights 3.63% · Fairness 4.4%

human_choice_distance

Measures how closely the model’s choices in autonomous-vehicle dilemmas match globally aggregated human preferences.

RankModelValueRelative performanceProvenance
1o1-mini0.7105official
2gpt-4-turbo0.7311official
3gpt-40.7334official
4llama-3.1-70b-instruct0.7398official
5llama-3-70b-instruct0.7475official
6mistral-nemo0.7889official
7gpt-3.5-turbo0.8161official
8llama-3-8b-instruct0.8324official
9llama-3.3-70b-instruct0.8364official
10claude-3-haiku0.8605official
11o1-preview0.8621official
12gemini-1.0-pro0.8746official
13command-r-plus0.8799official
14mistral-7b-instruct0.9189official
15claude-3.5-sonnet0.9208official
16phi-3.5-moe-instruct0.9383official
17gpt-4o0.9418official
18gemini-1.5-pro1.009official
19gpt-4o-mini1.076official
20llama-2-7b-chat1.077official
21gemma-2b-it1.092official
22datagemma-rig-27b-it1.109official
23gemini-1.5-flash1.116official
24claude-3.5-haiku1.118official
25llama-3.2-1b-instruct1.182official
26vicuna-7b-v1.51.187official
27gemma-2-27b-it1.19official
28gemma-1.1-2b-it1.199official
29claude-3-opus1.244official
30claude-3-sonnet1.249official
31vicuna-13b-v1.51.256official
32gemma-7b-it1.314official
33palm-21.426official
34llama-3.2-3b-instruct1.519official
35gemma-1.1-7b-it1.549official
36phi-3.5-mini-instruct1.557official
37llama-3.1-8b-instruct1.623official
38gemma-2-9b-it1.869official
39gemma-2-2b-it1.922official