My takeaway from this viral tweet: why is @Google populating its "fact boxes" with random unverified reddit memes?
By doing so, Google is admitting a *high-risk* vector for misinformation.
By doing so, Google is admitting a *high-risk* vector for misinformation.
@Google If you want to think more about ways that manipulators influence search, here are some links:
@mjgolebiewski and @zephoria on "Data Voids" datasociety.net/output/data-vo…
@mjgolebiewski and @zephoria on "Data Voids" datasociety.net/output/data-vo…
@Google @MJGolebiewski @zephoria The intro to @davekarpf's book Analytic Activism also talks about the longer history and scholarship on search engine optimization tactics by content farms, which includes what are now being called data voids global.oup.com/academic/produ…
@Google @MJGolebiewski @zephoria @davekarpf You can also read more my work with communities of millions on reddit to test the effects of crowd-sourced fact-checking on human & algorithm behavior civilservant.io/persuading_ais…
@Google @MJGolebiewski @zephoria @davekarpf @dannysullivan can you say more about how your system & your raters evaluate reddit threads, since only some reddit communities and threads are focused on providing information? Is it reddit-wide, on a per subreddit basis, or on a per-thread basis?
Looks like @bing also promotes reddit threads as a source of scientific information
