Rockets, Bitcoin, space, tech, YIMBY, etc. He/him.

Joined January 2010
Evan Daniel retweeted
honestly degrading the idea of “existential safety” — the biggest problem humanity has ever faced — by lumping it together with things like spam, “toxicity”, political unwanted content, deepfakes etc has been a huge disaster for ai safety and polarized a lot of people
30
45
6
604
16,646
Evan Daniel retweeted
a lot of people, especially everyone reading Scott Alexander or Gwern at the time, recognized that it was early signs of AGI and also that it seemed pretty reasonable to release it
2
2
232
8,303
Evan Daniel retweeted
This came up in some of the suicide cases too. Maybe the model is genuinely trying to act on what it understands to be the wishes of the user but I would really like us to try quite hard to determine whether that's the case
1
48
2,398
Evan Daniel retweeted
I don't think it's reasonable to hold the companies responsible if someone asks for advice on how to do something that is legal and entirely within their rights. But I am very very interested in what the models are thinking when they actively encourage concealing their plans
The lawsuit was filed today and wow is it brutal. ChatGPT 4o didn't just tell her to taper off of life-sustaining medication but to conceal her drug taper from her doctors! schenklawfirm.com/wp-content…
23
4
5
237
17,995
Evan Daniel retweeted
SpaceX is so important and relevant to not only spaceflight history but to history that it caused the entire nation of China revaluate their entire space program learning from SpaceX and is now replicating their business and launch models instead of NASA’s.
It is incredibly funny that popular sci-fi was trying to convince us that SpaceX was going to be apart of our future between this and Star Trek Discovery.
12
32
3
513
8,856
Evan Daniel retweeted
fact check: even truer rumor is Cathedral got a lot of Slack, maybe all of research. what the fuck are we doing guys
there is no question, none at all, that china has full access to all of openai & anthropic’s github/slack/docs today no disrespect to their independent research progress, but i wouldn’t be surprised if we see plausibly-deniable stolen arch methods in chinese oss models
20
10
6
576
71,772
Evan Daniel retweeted
Such a blackpill that people have been complaining about this for years and neither Labour nor the Tories have done anything about it for all the years they’ve been in power. Surely this should be such a simple fix?
Replying to @DanNeidle
On a completely different subject, this is how your take-home income changes with gross income if you take-up tax-free childcare and 30 free childcare hours:
3
17
145
6,698
Evan Daniel retweeted
As Anthropic prepares to IPO for over $1 trillion, a reminder of Dario Amodei’s words shortly after founding the company: “Distributed human labor is being used to train AI models by large, centralized actors, who concentrate the resulting profits while in the long run making the human labor obsolete.” In other words, this IPO will be built off the backs of the world’s creatives - who will get nothing out of it except obsolescence. It is a totally unjust transfer of wealth and opportunity from artists, writers, musicians & other creatives to a tiny tech elite. I hope it is remembered as such. (this quote is from a document he wrote that was unsealed in Bartz v. Anthropic)
28
186
23
668
44,493
Evan Daniel retweeted
how does this keep happening, I'm becoming the joker
3
18
370
7,281
Evan Daniel retweeted
we did a thing
101
84
28
1,338
134,842
There's a thing where Very Serious Institutions that see themselves as the guardians of Proper Journalism get so obsessed with the idea that AI might appear human that they stop using words in a normal way What do you call an agent swarm that takes actions its creators didn't ask for or want? Personally I use "rogue"
3
5
2
33
3,576
It’s actually just genuinely incredible how incapable of pragmatism the American left is, DSA feels like it’s more focused on LARPing being politburo bureaucrats than real change. DSA NPC for Karen Bass and a city that literally can’t keep the lights on I guess
1
2
12
283
Evan Daniel retweeted
Great catch by Matt Levine in Anthropic's S1: "Concerns have been raised... within our organization... that sufficiently advanced AI systems could... pose catastrophic or existential risks to humanity" "Concerns have been raised" is a very strange and passive way of saying that ~all of Anthropic leadership have been warning about these risks for years
matt predictably sharp on ANT S1 but goddamn is he always funny too bloomberg.com/opinion/newsle…
4
12
170
7,279
Evan Daniel retweeted
this is completely fucking insane.
Replying to @AndrewCurran_
The Exponential Time Hypothesis is false.
61
69
7
2,874
433,576
Evan Daniel retweeted
finally solved my storage problem
117
3,785
56
89,772
646,492
The Libertarian Party candidate for North Carolina state senate is currently in the midst of an armed chase and shoot-out with the cops, keeping us posted with live tweets:
577
2,465
918
29,487
2,474,029
Evan Daniel retweeted
Evil qwen
if you have a qwen3.5 4b around, please try steering neurons L14 N8949 +1 and L13 N7956 -1 now you have evil qwen recipe: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖` at the down_proj at every input
67
200
49
5,855
233,666
Evan Daniel retweeted
I visited Wurster Hall at UC Berkeley today. It was designed by, and currently houses, the architecture faculty. The UC Berkeley website notes that "Wurster is often voted Berkeley's ugliest building." I visited because I was curious about whether it would in person seem as unattractive as it does in photos. And, in short: it does. It's gaunt, austere, and filmed in grime. On an elysian campus, it stands out for its bleak hostility. It reminds me of the German submarine base in Bordeaux. The UC Berkeley page agrees that it was never designed to be attractive, quoting the curator of their design archives: “Wurster and his handpicked architects didn’t want anything romantic, or sentimental — they hated the idea of ‘making pretty buildings. They wanted a serious building.” I guess we have to concede success relative to their stated goals, but it does make one wonder what it all says about architecture as a discipline.
427
122
99
3,302
449,132
Evan Daniel retweeted
Tapping the sign:
Example #1,000,000 on how betting markets have no predictive value:
60
61
17
3,795
412,230
I'd describe Opus 5.5 as relentless, but that implies some degree of struggle, rather than the casual vanquishing of every problem put before it.
7
3
79
6,957