VP at NRG | Public opinion researcher | Former @washingtonpost political data journalist | Dad, people person, math person!
- Tweets 19,300
- Following 810
- Followers 16,610
- Likes 11,148
"Escaped crickets while feeding the iguana in the office" day sounds like the _best_ work day
1
5
414
This was from late August (prior to the current round of AI doom conversation)
But the public basically gets their AI info via scroll, while looking for other stuff. Which seems significant!
(US adults age 18-64)
1
685
I wrote a little tidbit on this today, might expand more bc it's important and I have more substack.com/home/post/p-217…
1
448
half of democrats having never heard of hasan piker vs only 13% for bill maher is really a throwback "twitter is not real life" finding
Favorability Ratings among Democrats
Obama: 90/4 (+86)
Buttigieg: 70/7 (+63)
Sanders: 72/15 (+57)
Mr. Rogers: 61/6 (+55)
Harris: 70/20 (+50)
Ossoff: 55/5 (+50)
Kelly: 58/8 (+50)
Ocasio-Cortez: 62/15 (+47)
Newsom: 63/17 (+46)
Cooper: 58/12 (+46)
Mamdani: 57/12 (+45)
Shapiro: 55/10 (+45)
Beshear: 49/5 (+44)
Pritzker: 44/7 (+37)
Khanna: 33/6 (+27)
Jeffries: 45/22 (+23)
Fain: 13/7 (+6)
Schumer: 35/32 (+3)
Hassan Piker: 14/14 (+0)
Bill Maher: 27/34 (-7)
Center for Strategic Politics | 9/3-10
liberalcurrents.com/what-are…
1
3
15
2,226
This is what I've seen (and talked about with @galendruke earlier this week!)
If you say "hey should we regulate AI on X sub-issue or try to prevent harm Y" people are mostly like "YEAH DO THAT"
x-risk hasn't been top of mind so far (as measured by what people bring up in OEs). But my last AI survey closed end of August. Currently planning to check again later this month.
Yeah, I think you're likely to see very large and relatively bipartisan majorities concerned about AI risk. People will have many different reasons for these concerns, some more well-grounded than others, but likely a fair amount of directional agreement.
2
2
996
Really enjoyed this! Two other topics we talked through that were of especially fun/interesting:
-- Fuzziness level of AI issues: data centers are the only concrete, present-tense issue that's landed w/the public. job loss is hypothetical, doom is elite discourse, etc.
-- Ways in which asking people OEs about what they're worried about vs a list elicits similar results, but different enough to tell us real things
There's a lot more! But those are in my head rn
I'll say it: P(doom) is bullshit.
@databyler puts it much more eloquently than me tho:
1
2,457
David Byler retweeted
Some professional news: after nearly ten years at The Washington Post, I'm super excited to have finally started something of my own: Broome Street Labs.
12
11
161
11,047
p(doom) is a take dressed up as a number
that's okay! not all takes are bad, and TBH you can't get much more empirical
but that is what it is
3
2
4
1,285
Thank you for highlighting!! Stay tuned, will have more on our our monthly poll soon :)
The most AI-exposed and least adaptive jobs are held by women -- to the tune of 80-90% women in some cases. Women don't like AI. They are using it less.
This week in my AI newsletter, I break down the data on how women are getting left behind - and keeping themselves behind.
Plus, more on Flock cameras and a nerdy methods bonus.
2
1,198
I feel like you could do something fun with an expert panel, X months prior to election -> see how predicted WAR lines up against results
We need a WAR system for first time candidates. Yes, I get we don't necessarily have an electoral track record, can instead be brutally honest personality assessments, etc.
1
2
1,491
there was a big public call for state-level primary polls, and those races are notoriously hard to poll 🤷
1
11
2,042