4 ms·

China Tests US Voter Fault Lines with AI Content

15 points by 2OEH8eoCRo0 3y ago

probably_jesus 3y ago
[flagged]
wumeow 3y ago
Posing a question at the end of a tweet like in figure 12 is a common Engagement 101 tactic to boost replies to a post and get it picked up by the algorithm.
beardyw 3y ago
> Chinese Communist Party (CCP)-affiliated actors have started to pose contentious questions on controversial U.S. domestic issues to better understand the key issues that divide U.S. voters Surely you can be buried under the weight of opinions and crackpot ideas without needing to ask.
- 082349872349872 3y ago
  > ...and crackpot ideas without needing to ask. The trouble with crackpot ideas is that they're closer to 5/95 splits, so only useful to the extent they tell you who's likely to spread other crackpot ideas. It is a good point with regard to 50/50 splits (maybe even 40/60 if they correlate with the gerrymander?), but the issue you correctly point out is that these are all well known, as they've been adequately researched and publicised by the the indigenous political parties. I've heard that some fast food franchises don't even bother doing their own market research: if there's a McDo on a corner, they'll try to get one of the remaining three locations. Following domestic faults would be a similar high-payoff/incredibly-low-effort strategy. EDIT: compare "On Trolling": https://news.ycombinator.com/item?id=23830829 https://news.ycombinator.com/item?id=23830829 (come to think of it, I wonder if the "posit the not-quite-controversial" troll strategy mentioned there could function like dithering for posterised images?)
082349872349872 3y ago
As if US political parties aren't already doing sufficient probing of voter fault lines? (I wonder how much the CCP would be willing to sell their data for?) In any case, the real information is not what aggregates respond to Sort By Controversial content, but what the specific pattern of cliques that forward it may be. In the XX such information was hand-gathered, but now I wonder how much Meta et.al. would be willing to sell their data for?

15 points by 2OEH8eoCRo0 3y ago

[flagged]

Posing a question at the end of a tweet like in figure 12 is a common Engagement 101 tactic to boost replies to a post and get it picked up by the algorithm.

beardyw 3y ago

> Chinese Communist Party (CCP)-affiliated actors have started to pose contentious questions on controversial U.S. domestic issues to better understand the key issues that divide U.S. voters Surely you can be buried under the weight of opinions and crackpot ideas without needing to ask.

082349872349872 3y ago
> ...and crackpot ideas without needing to ask. The trouble with crackpot ideas is that they're closer to 5/95 splits, so only useful to the extent they tell you who's likely to spread other crackpot ideas. It is a good point with regard to 50/50 splits (maybe even 40/60 if they correlate with the gerrymander?), but the issue you correctly point out is that these are all well known, as they've been adequately researched and publicised by the the indigenous political parties. I've heard that some fast food franchises don't even bother doing their own market research: if there's a McDo on a corner, they'll try to get one of the remaining three locations. Following domestic faults would be a similar high-payoff/incredibly-low-effort strategy. EDIT: compare "On Trolling": https://news.ycombinator.com/item?id=23830829 https://news.ycombinator.com/item?id=23830829 (come to think of it, I wonder if the "posit the not-quite-controversial" troll strategy mentioned there could function like dithering for posterised images?)

082349872349872 3y ago

As if US political parties aren't already doing sufficient probing of voter fault lines? (I wonder how much the CCP would be willing to sell their data for?) In any case, the real information is not what aggregates respond to Sort By Controversial content, but what the specific pattern of cliques that forward it may be. In the XX such information was hand-gathered, but now I wonder how much Meta et.al. would be willing to sell their data for?