Luxembourgish parliament passes AI code of ethics

Aligned with the recently passed EU Regulation on Artificial Intelligence, the AI code of ethics offers ten non-binding guidelines that seek to clarify the internal use of AI for parliamentarians and parliamentary staff.

RTL's Fanny Kinsch spoke with Laurent Scheeck, General Secretary of the Parliament on the use of AI in parliament.

Four years ago, the parliament began working on a framework on the use of AI, according to Laurent Scheeck. He underlined the importance of having a set of ethical guidelines on the use of AI, noting particularly that AI should not be a substitute for humans.

"That is categorically out of the question. This is why, with this code of ethics, we want to be as transparent as possible and ease people's worries."

The code of ethics affirms that AI should only be used in a supportive capacity. It aims to help parliamentarians and staff better understand and navigate the implications of using AI, stipulating that a risk analysis must be conducted prior to any use of AI.

"The goal here is to fully leverage the possibilities offered by AI while alleviating any fears."
 
The code does not cover how AI should be used for political work, instead it is primarily aimed at parliamentary staff. Currently, AI is being used in three areas of parliamentary work: transcribing, archiving and translating.

"We have made great progress in transcribing parliamentary debates, where we are already using a tool developed in collaboration with the University of Luxembourg."
 
Scheeck shared that the Parliament hopes to work with the university again in the field of translation. 
 
"One of our priorities is to provide subtitles in the future, to better reflect the linguistic pluralism of our country."

One of the ways in which AI is used in archiving, is with AI's ability to identify sensitive data. This would help staff categorise and analyse documents more efficiently.