ჯუნგლების ცხოვრება

397 posts

ჯუნგლების ცხოვრება

ჯუნგლების ცხოვრება

@Anthurium_xanth

Katılım Nisan 2025
587 Takip Edilen18 Takipçiler
ჯუნგლების ცხოვრება retweetledi
Sridhar Ramesh
Sridhar Ramesh@RadishHarmers·
You racists are all pathetic. Trump won't make your life better in any way, and once he is gone you will have nothing. You are not going to be the celebrated figures of even the near future.
English
0
46
1.2K
19.9K
ჯუნგლების ცხოვრება retweetledi
Ethan Mollick
Ethan Mollick@emollick·
I've been going on about how ChatGPT Work (and Cowork) are missed opportunities for knowledge workers, and to illustrate that take a look at Google's NotebookLM answering the same question as ChatGPT Work, with the same 70+ files. It centers process & sources, not just outputs.
Ethan Mollick tweet mediaEthan Mollick tweet mediaEthan Mollick tweet media
English
67
46
925
192.2K
ჯუნგლების ცხოვრება
@roanoke_gal @ramez I am Australian and we do not have a police state. As per my comment to the OP, most people on Earth simply don't share your moral intuitions so you're going to have to actually make an argument rather than gesturing to things that don't exist, such as Australian police states.
English
0
0
1
15
roanoke_gal
roanoke_gal@roanoke_gal·
it doesn't. but the alternative (preventing it) is much worse, as it inevitably leads to authoritarian regimes banning open source/weight models, and keeping closed source ones either completely restricted ala mythos, or heavily nerfed/censored. meanwhile those in power can & will continue using the unrestricted versions to fully cement their power. like guns - of course its terrible that people use them to kill loved ones. but banning them leads to *gestures at the growing police states in the uk & australia*
English
1
0
0
11
roanoke_gal
roanoke_gal@roanoke_gal·
oh no i meant the us government using claude & aiming at obvious military targets like *checks* girls' elementary schools. but more generally, i'm saying that a world where any governments can use freely uncensored & unfiltered ai's while ordinary people are hamstrung by "safety" inevitably leads to a world where the epstein class will rule, free of consequence
English
1
0
3
60
Ramez Naam
Ramez Naam@ramez·
@Anthurium_xanth I believe that government should have the same power to restrict AI content as they do to restrict what can be published on the internet or in books. And that is nearly zero.
English
1
0
1
16
roanoke_gal
roanoke_gal@roanoke_gal·
they're already letting the government commit mass-murder in iran - i'd much rather they not get to keep all the killbots to themselves, that way lies a future much more terrifying than one where some rando occasionally gets murder advice from chatty.
English
1
0
0
80
Ramez Naam
Ramez Naam@ramez·
@Anthurium_xanth That's the unhinged part. And it is darkly amusing. And it's correct from a civil liberties perspective.
English
1
0
3
202
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
@Anthurium_xanth That's why I recommend countries find way to be useful to the stack, because, just like with national defense, "they will spend money and incur security risk to help us just because" is not a good strategy. Note Australia also doesn't make AI chips, memory, many rare earths, etc
English
1
0
0
20
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
Today in "US is not unique, nor is closed-source; no country will allow unrestricted use and export of frontier dual use intelligence models". Can't stress enough for middle powers that "ensure access to frontier AI" is going to be more important than "be sovereign from US/China"
Jukan@jukan05

CHINA CONSIDERS RESTRICTING OVERSEAS ACCESS TO CUTTING-EDGE AI MODELS China’s Ministry of Commerce has led meetings over the past month with major AI companies, including Alibaba, ByteDance, and Z.ai, to discuss measures that would restrict overseas access to cutting-edge AI models, including models that have not yet been released. The discussions reportedly include not only closed-source models but also open-weight models. However, the scope of application is still under debate, and the rules may ultimately apply only to future frontier models. Officials have also discussed designating the leakage or theft of proprietary AI technologies as a national security crime, with stronger penalties, as well as restricting the types of foreign capital that can invest in Chinese AI startups. The backdrop is the U.S. move to strengthen export controls on AI models, along with national security concerns over cutting-edge models that could possess advanced cyberattack capabilities. Chinese authorities are reportedly concerned that advanced U.S. cybersecurity AI models could be used to exploit vulnerabilities in Chinese software. Since the beginning of this year, China has continued to tighten measures to prevent AI technology from being transferred overseas. Authorities have investigated whether Chinese AI startups that relocated abroad violated export control laws, while also strengthening oversight of overseas transactions involving Chinese investors, technology, data, and national security concerns. Future regulations could take the form of a tiered framework based on technological capability. Basic open-source AI models may be managed through a filing system, high-performance models may be subject to security reviews, and the most sensitive frontier models may be banned from public release or restricted to use within China.

English
2
3
24
3K
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
@Anthurium_xanth Not really! You just use the same raw materials/$/whatever as leverage in the AI stack as you do with these things as leverage in the defense and other security stacks. It isn't rocket science.
English
1
0
0
15
Kevin A. Bryan
Kevin A. Bryan@Afinetheorem·
@Anthurium_xanth 1) No one outside US or China are even trying to make frontier AI. 2) And this means distilling second-tier even harder. 3) Access to frontier AI therefore requires agreements with US or China or both. 4) You need leverage + diplomacy.
English
1
0
0
21
Arin Dube
Arin Dube@arindube·
Actively unfollowing some prominent AI boosters tonight.
English
2
0
33
7.4K
Kashiyuka fan
Kashiyuka fan@hndlehndlekyun·
@alz_zyd_ I agree in principle but I suspect Vance is interested neither in reading your tweets nor learning about politically inconvenient ideas.
English
1
0
3
85
alz
alz@alz_zyd_·
I've never been a fan of this approach. If someone has questions about what we do, we should just answer the questions! In tweet length! "just read MWG, dumbass" is... powertripping. Analogous to a priest saying "you would know this if you read the scriptures as much as I did"
alz tweet media
Stéphane Surprenant@S_Surprenant

If you want to make claims about what economists think, you probably should be reading what we write. This isn't a matter of it being economics, but a matter of data collection: you need a representative sample of what we say to make statements about what we say. If you aren't trying to make sense of a large number of papers and you're saying things like "economists think X," you aren't "thinking for yourself." You're just making shit up. And that's the problem with JD Vance's comments. He is making shit up. More generally, if you want to think about economic measurement, you should read work on the topic to understand how it works, why it was done this way, and to see how it is used in practice. If you have not spent even a couple of hours on this, how would you know if there are problems? You don't even know what you're looking at. But if you did your readings, maybe some things do feel odd to you and you might then ask if that idea also occurred to someone else. There are people who obsess over problems like those full time, so the answer is often "yes." Otherwise, that's a valid research topic -- someone could write a paper about it. If you have the skills, or a lot of time and the inclination, you could write that paper. Then you'll be the someone who thought about it next time someone else asks.

English
6
2
32
6.5K
Sam Bowman
Sam Bowman@s8mb·
I don't agree with the attacks on individuals in this post, but I think this is a pretty important point about how Twitter YIMBYism risks becoming detatched from reality about how easy it is to steamroll opponents of development and dismiss their objections. I didn't realise that there had been a backlash to @dc_lawrence's post the other day, which I thought made good points, and a lot of it seems to be based on a fantasy that you can just choose to dismiss the opinions of a large majority of voters and "just pass some laws" that they hate. I think that's obviously mistaken, and even though I think a lot of the arguments made against new development in the UK *are* actually pretexts for other objections, these other objections are still extant and require solutions. freddiedeboer.substack.com/p/do-the-yimby…
Sam Bowman tweet media
English
9
6
96
16.9K
Jeremiah Johnson 🌐
Jeremiah Johnson 🌐@JeremiahDJohns·
@s8mb I stand behind my posts, but I'm mostly confused why Freddie put my name in scare quotes.
English
5
0
25
1.6K
ჯუნგლების ცხოვრება retweetledi
Jack Meyer 🏛️
Jack Meyer 🏛️@Jackbmeyer·
To the extent that critics of economics overlook the purpose of assumptions in modeling, defenders can make the same mistake in their deference. The critique in question, with respect to quality and price, is certainly subject to vast amounts of economic research, but quality is nevertheless treated as an assumption behind many of the headline figures because the aggregates are largely accurate and useful in spite of this heterogeneity. However, the persistent gap between public sentiment and macroeconomic indicators suggests that examining where the common assumptions diverge from economic reality may be a worthwhile line of inquiry to pursue, even when a relevant literature on this topic already exists.
Jeremiah Johnson 🌐@JeremiahDJohns

There's a common type of criticism of economics where people think up a reasonable objection - like Vance saying prices ignore important factors like quality and other things - and then assume economists have never thought of it before. It's more likely that *thousands* of economists have done detailed research on the exact subject you're accusing economics of ignoring. I promise you that you are not the first person to have dreamed up this particular objection. People much smarter than you have already thought of it, collected data, built models to explore it, and written about the whole thing exhaustively.

English
1
4
41
3K
Michael Fowlie
Michael Fowlie@mwfowlie·
@alz_zyd_ You’re not wrong but almost every metric people care about is highly correlated with real GDP per capita. Note not real GDP, or nominal GDP per capita. That means education, healthcare, life expectancy, crime, and many more things. It’s a very good proxy for wellbeing.
English
2
2
75
2.6K
alz
alz@alz_zyd_·
I actually think Vance is very much on point here. Focusing on GDP too much as a sole metric of a society's success is misguided. There are many different numbers - fertility, health, military power, etc - and each has something different to say
alz tweet media
English
26
2
119
224.7K
Andrey Mir
Andrey Mir@Andrey4Mir·
Most of us think in McLuhan’s terms without even realizing it. For example, it was hard to analyze the effects of television separately from the content of TV shows. But now nearly everyone sees that social media polarize users regardless of the specific content circulating there. It’s the algorithmic maximization of engagement, not the content, that drives polarization. That’s what “the medium is the message” means. 1/3
Andrey Mir tweet media
English
2
11
38
14K