Random triples
| Subject | Object |
|---|---|
| gptkb:Claude_1 | AI alignment |
| gptkb:Claude | honesty |
| gptkb:Claude_2.0 | constitutional AI |
| gptkb:Claude_2.1 | high |
| gptkb:Claude_3_family | harmlessness |
| gptkb:Claude_3_family | transparency |
| gptkb:Claude_3_family | reduced bias |
| gptkb:Claude_model | harmlessness |
| gptkb:Med-PaLM_2 | bias reduction |
| gptkb:Claude_3_family | AI alignment |
| gptkb:Claude_model | helpfulness |
| gptkb:Claude_model | AI alignment |
| gptkb:Med-PaLM_2 | medical accuracy |
| gptkb:Claude_model | honesty |
| gptkb:Claude | AI alignment |
| gptkb:Med-PaLM_2 | explainability |
| gptkb:Claude_2 | high |
| gptkb:Claude | harmlessness |
| gptkb:Claude | helpfulness |