true hierarchy of needs moment
BREAKING: Krispy Kreme officially outperforms NVIDIA over the past year.View quoted post on X
Zac Stern
Verified account This account is verified. Learn more@sternzac
Founder of Undisclosed Group, Co-Founder of Golden Egg Media
Ex-Founder of @GetOfficialApp (acquired by @bumble)
zac@socialintelligencelabs.com
Ex-Founder of @GetOfficialApp (acquired by @bumble)
zac@socialintelligencelabs.com
948 Following 2.8K Followers 1.8K Posts
Latest posts 5
I bet $10,000 of my own money on the midterm results.
Planetarium, our newest simulation model at Social Intelligence Labs, ran a forecast of all 646 contests on the 2026 ballot.
So, after one very successful forecast on Amazon Prime Day, I thought we’d up the stakes.
Here’s Planetarium’s call:
-Senate: Democrats 68% (median 52–48)
-House: Democrats 95.1%
-Governors: D favored in 21 of 36
-Ballot measures: 106 of 145 favored to pass, 20 too close to call
Planetarium is built on 20,000 simulated elections, using simulated voter groups in every state. Using all publicly available data, we blended the polls where they exist and each state’s real counting rules. When we don’t have polls (2/3rds of races), we fully simulate the election.
For now, explore the Planetarium Midterms site, which details every single state’s race for governor and senator. askplanetarium.com/midterms
Comment on what you think I should spend my winnings on
Planetarium, our newest simulation model at Social Intelligence Labs, ran a forecast of all 646 contests on the 2026 ballot.
So, after one very successful forecast on Amazon Prime Day, I thought we’d up the stakes.
Here’s Planetarium’s call:
-Senate: Democrats 68% (median 52–48)
-House: Democrats 95.1%
-Governors: D favored in 21 of 36
-Ballot measures: 106 of 145 favored to pass, 20 too close to call
Planetarium is built on 20,000 simulated elections, using simulated voter groups in every state. Using all publicly available data, we blended the polls where they exist and each state’s real counting rules. When we don’t have polls (2/3rds of races), we fully simulate the election.
For now, explore the Planetarium Midterms site, which details every single state’s race for governor and senator. askplanetarium.com/midterms
Comment on what you think I should spend my winnings on
time for my hottest of all hot takes...
the whole movie is about manipulating the viewer to feel bad for her. but she went to prison for a reason. serious fraud, ppl were hurt
went in wanting to see the nathan fielder i know and love, but we can't all get what we want
the whole movie is about manipulating the viewer to feel bad for her. but she went to prison for a reason. serious fraud, ppl were hurt
went in wanting to see the nathan fielder i know and love, but we can't all get what we want
‘YOU CAN SEE EVERYTHING’ is the highest-rated feature film of all time on Letterboxd at 4.8.View quoted post on X
Turns out we can predict the future
excited for what comes next
excited for what comes next
Our prediction for this Prime Day comes from Mimi, our newest prediction model at Social Intelligence Labs.
Mimi simulates 60 million people using census data and enriched persona data from other public sources and social media. We have been building Mimi over the past few months and are excited to test it publicly for the first time.
Overall, here are our thoughts:
A smaller event than Prime Day. About $109 per household, roughly a quarter below June's $143, with orders around $47 (June: $48). About half of households place two or more orders (June: 63%).
Awareness high, enthusiasm modest. 91% knew about the event, but only about one in three (32%) shopped mainly because of it, and 56% are very or extremely satisfied with the deals.
Careful shoppers. About one in three (32%) are more cautious because of the economy, and 54% compare prices across
retailers.
Early holiday shopping, light AI use. About one in four (23%) buy holiday gifts; about one in five (21%) use AI shopping tools
We will report back when Numerator's data comes out in a couple of weeks.View quoted post on X
hmmm not sure i’d call this almost as good as chatgpt
muse is still losing a pretty significant chunk of users between day 7 and 14
seems like a lot of initial curiosity that isn’t turning into regular usage. retention should be a flat and steady line.
muse is still losing a pretty significant chunk of users between day 7 and 14
seems like a lot of initial curiosity that isn’t turning into regular usage. retention should be a flat and steady line.
Meta Muse's retention is out, and it's almost as good as ChatGPT!
It's higher than ChatGPT at D7 (>30%) and just lower at D14 (22%). Not quite in the elite D14 tier of WA (67%) YT (59%) Insta (54%) or TikTok/Snap (37%), though.
Muse is one of the highest retaining AI products!View quoted post on X
