Talk:Trustworthy AI

create stub on United Nations Statistics Division and ITU (UN ICT Agency) joint progrm on Privacy enhancing technologie standardization
Including homomorphic encryption, federated learning, and others Bquast (talk) 21:12, 24 October 2022 (UTC)

COI tag (November 2022)
ITU program, being primarily edited by ITU employee MrOllie (talk) 19:10, 26 November 2022 (UTC)

/!\ Homonymy with the Trustworthy AI as introduced in 2019 by the European Commission
An homonyms distinction page shall be created as Trustworhy AI refers to the concept of ethics in AI as developed by the European Commission HLEG AI (independent High Level Expert Group on Artificial Intelligence) since 2019 in at least 4 official documents which all prelude the European IA Act regulation :

1) AI HLEG - Ethics guidelines for trustworthy AI: https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=60419

2) AI HLEG - Policy and investment recommendations for trustworthy Artificial Intelligence: https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=6034-3

3) AI HLEG - Assessment List for Trustworthy Artificial Intelligence (ALTAI): https://ec.europa.eu/newsroom/dae/document.cfm?doc_id=68342

4) AI HLEG - Sectoral Considerations on the Policy and Investment Recommendations for Trustworthy AI: https://futurium.ec.europa.eu/sites/default/files/2020-07/Sectoral%20Considerations%20On%20The%20Policy%20And%20Investment%20Recommendations%20For%20Trustworthy%20Artificial%20Intelligence_0.pdf

Such other "Trustworthy AI" page is still to be created (done in French) and linked to all other pages about ethics for AI. Golem Le Brut (talk) 16:33, 9 May 2023 (UTC)

Made the article more generic
I changed the article so that it presents what "trustworthy AI" is in general, rather than the ITU programme in particular. I'm not sure whether the ITU programme is notable enough to have its own article, for now it's in one section of the article.

One thing that may still need to change is that the article is very focused on data privacy and standardization, which is what the ITU focuses on, but trustworthy AI in general seems to encompass more than that. Alenoach (talk) 12:37, 1 June 2024 (UTC)