Abstract
“Loki Test” challenges AI obedience, with a chatbot named after the god of mischief aiming to joke rather than being satisfied and pleased. It sharply reflects the bias in AI's database and encourages critical thinking about AI trustworthiness. Unlike obedient AI assistants on the market, Loki prompts caution in trusting AI.
| Original language | English |
|---|---|
| Publication status | Published - 23 Jun 2024 |
| Event | 2nd international workshop on eXplainable AI for the Arts at the ACM Creativity and Cognition Conference 2024 - hybrid (in-person and online), Chicago, United States Duration: 23 Jun 2024 → 23 Jun 2024 https://xaixarts.github.io/2024.html |
Workshop
| Workshop | 2nd international workshop on eXplainable AI for the Arts at the ACM Creativity and Cognition Conference 2024 |
|---|---|
| Abbreviated title | XAIxArts 2024 |
| Country/Territory | United States |
| City | Chicago |
| Period | 23/06/2024 → 23/06/2024 |
| Internet address |