wikipedia

The three principles that drive the Foundation’s deployment of machine learning solutions in Wikimedia projects are sustainability, equity, and transparency.

The key is that knowledge is a human endeavor, and the development of Wikipedia and its sister projects depends on human collaboration and consensus. Machine learning tools can augment the process, but AI cannot replace volunteer Wikimedians.

Welcoming language, customized wayfinding, instructional onboarding, familiar contribution and consumption interfaces, suggested actions, and explicit gratitude and feedback are all areas where small changes could make significant improvements in the Wikimedia experience—and potentially improve the quality of new editor contributions and overall re... See more

they view the encyclopedia as a “consensus truth,” as one of them put it: It acts as a reality check in a society where facts are increasingly contested. That truth is less about data points — “How old is Joe Biden?” — than about complex events like the Covid-19 pandemic, in which facts are constantly evolving, frequently distorted and furiously de... See more

“Sharing knowledge is by nature an act of kindness,”

“We’ve had artificial-intelligence tools and bots since 2002, and we’ve had a team dedicated to machine learning since 2017,” Selena Deckelmann, Wikimedia’s chief technology officer, told me. “They’re extremely valuable for semiautomated content review, and especially for translations.

Wikipedia is probably the most important single source in the training of A.I. models

Some Wikipedians remark that their endeavor works in practice, but not in theory.

Over the past decade it has become a kind of factual netting that holds the whole digital world together.