Source |
Wired Threat Level |
Identifiant |
3577602 |
Date de publication |
2021-10-28 11:00:00 (vue: 2021-10-28 23:06:14) |
Titre |
This Program Can Give AI a Sense of Ethics-Sometimes |
Texte |
Researchers trained an algorithm to answer questions about human values. Some of the responses are troubling. |
Envoyé |
Oui |
Condensat |
about algorithm answer are can ethics give human program questions researchers responses sense some sometimes trained troubling values |
Tags |
|
Stories |
|
Notes |
|
Move |
|