Nate Silver
@NateSilver538
Followers
3M
Following
9K
Media
4K
Statuses
44K
New Book, On The Edge, #5 NYT Bestseller! https://t.co/WeCLEOd4Be
New York
Joined August 2008
It may be an off-year in the election cycle, but don’t call it boring! Grab a drink and join @NateSilver538, @ClareMalone and me on November 3rd at the @ComedyCellarUSA in NYC as we make sense of this year's contests just 24 hours before polls close: https://t.co/hyPARp8nwy
gdpolitics.com
Join Nate Silver, Clare Malone, and Galen Druke for a live taping on November 3rd.
0
47
56
3. Not clear that "OpenAI could become the next Google/Meta as a consolation prize even if they don't achieve AGI on near timelines" is necessarily bad for valuations, especially since it's hard to figure out how stocks should price in a possibility of singularity + p(doom).
3
0
36
1. This is more about their perceived timelines than whether they're AGI-pilled (clearly yes) 2. What matters re: valuations is perceptions relative to the market. I thought the market was slow to recognize AI potential before. Not sure if erring in the opposite direction now.
I promise you that 'openai is secretly not agi-pilled' is a bad take if you believe it, I'd be excited to take the opposite side from you in a wide variety of financial transactions
2
3
67
Instead, they're loosening guardrails in a way that will probably raise more revenues and might attract more capital and/or justify current valuations. They might still be an extremely valuable company as the new Meta/Google/etc. But feels more like "AI as normal technology."
31
7
523
If you think the singularity is happening in 6-24 months, you preserve brand prestige to draw a more sympathetic reaction from regulators and attract/retain the best talent ... rather than getting into "erotica for verified adults."
12
16
618
Should save this for a newsletter, but OpenAI's recent actions don't seem to be consistent with a company that believes AGI is right around the corner.
We made ChatGPT pretty restrictive to make sure we were being careful with mental health issues. We realize this made it less useful/enjoyable to many users who had no mental health problems, but given the seriousness of the issue we wanted to get this right. Now that we have
98
153
3K
For our NFL team ratings, what would you rather see as a *default*? (FYI, you'll be able to toggle between these to compare them.) 1) Team ratings with adjustments for current injuries for the forthcoming week 2) Team ratings with healthy QB1 and other players
7
4
14
Weekly QBERT ratings. Drake Maye is quickly moving up the list. (FYI, this chart is *projected* ratings going forward, not the best so far this season.) https://t.co/jPzxIA51zG
20
50
117
And really, these considerations apply even before then. With teams also starting out with better field position with the new kickoff rules, they're often in a position where two first downs = field goal range. Don't love it, don't hate it, but it's definitely different.
8
0
66
The field goals thing is sort of related to it. If you're anywhere past the opponents' 45-yard line or so, it's riskier to gamble with downfield passes that may result in an INT because the expected value of the possession is higher when you're almost assured of a FG.
interesting that strategic optimization has pushed basketball and football in opposite directions: long shots vs. short passes NFL in 2025 has: - highest QB completion % ever - lowest INT% ever - fewest yards/catch ever - and, sort of a separate thing, but: most long field goal
12
9
232
They're usually worth "gambling" on. If it would take me 30 minutes to figure something out on my own, and 1) saves me 25 minutes, 2) saves me 15 minutes and 3) costs me 30 minutes, the expected value is positive. But there's still quite a bit of 3).
55
11
463
For low-complexity tasks, they're almost always helpful. For medium-complexity tasks, probably split about evenly between 1) works like magic on the first try; 2) wrong but directionally helpful or works after refinements, and 3) proves to be a net waste of time.
9
15
508
I'm just one person, and my programming needs are somewhat unusual (building various kinds of statistical forecasting models). But I'm just not seeing the consistent productivity gains from LLMs that I would have expected if you'd asked me 6 months ago.
200
115
2K
This has been a great game, but not *nearly* as great as if they'd randomly put a runner on second base at the start of every inning.
97
21
748
Neither progressiveses nor centrists dominate the Democratic Party. But the compromises haven't always been so great either. Here's my deep dive on left vs. center infighting over the past 50 years, primarily through the lens of presidential nominations.
29
20
163
From a NYT chat that published this morning, whole thing here: https://t.co/ABFsmvcYcu
nytimes.com
What the polling says, who’s up, who’s down — and when it might end.
3
1
20
The bigger thing, though, is that I'm not sure what Democrats' strategic objectives really are, beyond saving Schumer's job/boosting base morale. Tactially, it seems to be going pretty good! But tactics are not strategy.
12
5
55
I suppose I continue to view the shutdown differently from the CW. For one thing, I don't think the polling tells us anything particularly meaningful. Nor is it actually all that good for Dems, e.g. Trump approval hasn't declined. But mostly I just don't think it tells us much.
15
12
102