Cipolla

Karma: 16

Cipolla 25 May 2026 12:22 UTC
2 points
0
in reply to: Linch’s comment on: Linch’s Shortform
Is this language dependent?

Looking for backdoors in Jane Street LLMs

Cipolla23 May 2026 0:06 UTC

16 points

0 comments14 min readLW link

Cipolla 24 Feb 2026 22:47 UTC
1 point
0
in reply to: Richard_Ngo’s comment on: ricraz’s Shortform
Yes, there is reticence about lots topics, not just intelligence differences. But the fact that some people are unwilling to fully engage with the implications of that, I guess as an explanation of problems happening in society, does not make that point more valid.
Mental blocks generally happen when people avoid thinking about bad scenarios, when there is discomfort (e.g. not having enough money for X, or a plan going bad, or a bridge breaking, or social pressure). I guess the trick you did is removing some bad scenario that allowed you to further explore certain topics; this does not mean you are immune from ‘bad’ scenarios blocking you on that same topic (e.g. reversing to the mean of that topic).
As a personal observation, I keep noticing on this forum the frequency of your eloquent “Pindaric” flights to nudge conversation towards IQ. I wonder why. But I will refrain from commenting further, only suggesting that it might be good to travel around the world and to engage with real people.

Cipolla 4 Dec 2025 11:12 UTC
1 point
0
on: Cipolla’s Shortform
From https://sciencepolicy.colorado.edu/students/envs_5110/collins_the_golem.pdf, Introduction:
...Both these ideas of science are wrong and dangerous. The personality of science is neither that of a chivalrous knight nor that of a pitiless juggernaut. What, then, is science? Science is a golem.
A golem is a creature of Jewish mythology. It is a humanoid made by man from clay and water, with incantations and spells. It is powerful. It grows a little more powerful every day. It will follow orders, do your work, and protect you from the ever threatening enemy. But it is clumsy and dangerous. Without control, a golem may destroy its masters with its flailing vigour.
The idea of a golem takes on different connotations in different legends. In some the golem is terrifyingly evil, but there is a more homely tradition: in the Yiddish brought from the East European ghetto, a golem (pronounced ‘goilem’ in that dialect), is a metaphor...
And similarly, from https://xcelab.net/rmpubs/sr2/statisticalrethinking2_chapters1and2.pdf, Chapter 1 (bolding is mine):
A golem (goh-lem) is a clay robot from Jewish folklore, constructed from dust and fire and water. It is brought to life by inscribing emet, Hebrew for “truth,” on its brow. Animated by truth, but lacking free will, a golem always does exactly what it is told. This is lucky, because the golem is incredibly powerful, able to withstand and accomplish more than its creators could. However, its obedience also brings danger, as careless instructions or unexpected events can turn a golem against its makers. Its abundance of power is matched by its lack of wisdom. In some versions of the golem legend, Rabbi Judah Loew ben Bezalel sought a way to defend the Jews of Prague. As in many parts of 16th century Central Europe, the Jews of Prague were persecuted. Using secret techniques from the Kabbalah, Rabbi Judah was able to build a golem, animate it with “truth,” and order it to defend the Jewish people of Prague. Not everyone agreed with Judah’s action, fearing unintended consequences of toying with the power of life. Ultimately Judah was forced to destroy the golem, as its combination of extraordinary power with clumsiness eventually led to innocent deaths. Wiping away one letter from the inscription emet to spell instead met, “death,” Rabbi Judah decommissioned the robot.

Cipolla 19 Nov 2025 13:00 UTC
0 points
−1
on: Cipolla’s Shortform
If a machine passes the standard Turing test, this might be enough to take over humans, even if the machine is not aware of this. So, in principle you do not even need super-intelligence.

Cipolla 18 Nov 2025 20:02 UTC
1 point
0
in reply to: Buck’s comment on: GradientDissenter’s Shortform
I think you wanted to say standard error of the mean.

Cipolla 24 Oct 2025 13:56 UTC
1 point
0
in reply to: Noah Birnbaum’s comment on: Daniel Birnbaum’s Shortform
I had similar thoughts. And indeed I found funny that a forum of rationalists falls for this.
But this is human nature, and we have finite time and energy to focus on things.
You will find the same for the attention papers written by people at famous institutions get, even if low quality.
Making names, recently, I had a similar feeling when looking at “AI as Normal Technology”. I did not know the authors, and I was skimming it because someone talked about it.
It is also a reason for which sometimes celebrities are called to do propaganda.
A strategy would be to make make posts anonymous for a limited amount of time (perhaps with some text randomization that would slightly change the style of the author/s, even though this might affect quality).

Cipolla 25 Sep 2025 9:35 UTC
3 points
1
in reply to: Charbel-Raphaël’s comment on: Global Call for AI Red Lines—Signed by Nobel Laureates, Former Heads of State, and 200+ Prominent Figures
First of all, I appreciate your energy and time, and I am not saying all of this is not useful.
Second, I have a feeling that on this forum there is too much rationality going on, and less talk about random chance, human behavior, and so on.
Said this.
Point one
I am not sure how these surveys were done, but if they were done in a naive way they might be wrong (e.g. election polls are biased if you do not frame well questions).
And even if they were true and many people say they are against AI, they probably use it indirectly, as they are consumers of other services.
My point is that people, pro- and against- AI, are using products that are exploiting the power of AI in some way, and hence they are fueling incentives in expanding AI capabilities/applications.
Even DuckDuckgo is using AI. There are too many incentives everywhere to use it.
It is on the end user to deny the application of AI to their everyday life.
Point two
We need to look at this from multiple perspectives. And I tell you why I think this is different wrt biological or nuclear weapons.
AI is becoming/sourcing addiction around people. People are addicted to ChatGPT (e.g. it represents some safe space to them), or to social media scrolling. Go around on trains and buses. Everybody is on their phone. People lost ability to get bored! Moms are not looking anymore at their babies. Friends are constantly interacting on their phone, and AI will become their friend. Crazy.
So, once people become addicted to it, it will be more difficult, on average, to get away from it. And they will fuel more and more the process of self-improving AI systems. They will give big companies opportunities to do work. This attracts money. And money attracts talent (I mean, I would also like to work on something fun that makes me money, and forget about consequences on the very long term).
I would say this sort of addiction is similar to the alcohol one. It is border line. Did the ban of alcohol in US work? Will the ban on AI work? This will depend on the people. If the base is borderline addicted to it, is difficult to do anything.
This is one of the cases that you can not really change someone if they do not change themselves within.
All of this process will come back to Point 1: fueling economical incentives in a greedy system. We need to come back to first principles, and teach people basic principles.
Unfortunately, our current society is one of low self-consciousness, easy to manipulate and steer. We really think we are smarter and better than our ancestors, but truth is we are not.

Cipolla 25 Sep 2025 7:56 UTC
−2 points
−3
on: Global Call for AI Red Lines—Signed by Nobel Laureates, Former Heads of State, and 200+ Prominent Figures
Being a Nobel prize, head of state, or holding a prestigious stance in society does not mean you understand what you are talking about. And I do not trust much UN (basically a mafia organization that does lots of signaling work and does not even represent the human population).
Also, these international treaties were not proved to work or to be robust to emergency cases.
If it comes, it need to come from the streets and the homes. If random people tomorrow drop AI, I guarantee you things will change. You need to focus on the basics, with a bottom-up approach.

Cipolla 24 Sep 2025 16:40 UTC
0 points
0
on: Cipolla’s Shortform
I had a look at “AI as Normal Technology”, and some of the other writing of its authors. I thought it was two amateurs, until I checked their backgrounds. Astonishing how AI space is so random.
If you want a serious read check out for example Energy and Civilization: A History, by Vaclav Smil.

Cipolla 15 Sep 2025 11:55 UTC
3 points
0
on: Does My Appearance Primarily Matter for a Romantic Partner?
A little bit. But the most important aspects for attraction are the energy and feelings you transmit. I can list so many situations of ugly or poor bad dressed guys dating nice ladies (both at uni and outside).
Also, if you work on your appearance too much you may end up with false positives that waste your time, costing you on the long run; some people focus on status (be partner’s looks/money/position/etc...) and if you signal that you might not end up in a nice relationship.
In the long term you need to focus on values, direction in life, and self-confidence. Hitting the gym, having basic grooms is important, but just keep good health and good manners.
In my opinion.
P.S.
I think by the end of the day dating is a bit random.
During uni I had lots of pimples on my face, low self-confidence, but somehow I was attracting beautiful ladies that my friends, well-dressed and behaved, did not.
I heard of normal people having just a high-school degree with PhD girlfriends.
etc....
P.P.S.
Another thing. Focus one person by one person. You really need to be present in the situation. I heard so many people talking about how dating multiple ladies improve relaxation and confidence. But here is the thing. You, and the lady, need to be present in the moment. Like, there is no one else on Earth. Basic steps.
P.P.P.S.
Last thought. In the past, I had ladies rejecting me and changing their minds some time after. So, if you get rejected embrace it and have courage :)
P.P.P.P.S.
If you really have trouble/want alternative ways of meeting people, getting introduced is one of the best ways actually. So, through friends of friends/acquaintances, hobby groups.
You could also use Apps, I have several friends that met their girlfriends there, and the general advice is to show your activities. But I really think real life is the best and most organic way for finding a lady, statistically speaking.

Cipolla 9 Sep 2025 17:50 UTC
−4 points
0
on: Cipolla’s Shortform
I see no difference between the principles underlying a financial system that builds products to please a few, and a social system that allows anybody to be anything to please a few.
Both serve a relatively small number of people. Both are based on a kind of freedom of expression that puts the individual in the center. One has been proven to be a very unreliable system that has been leading to crises and human suffering. The other one, will just be a matter of time.
There is a proliferation of too much freedom, meaning no freedom.
I think most of the people on LW can not see it. I believe the reason is due to social class. The average poster on LW comes from a relatively privileged economic and social environment that has led them in positions to rule over the others.
The consequence is that few people, mainly due to luck (for a random event that happened in the past), end up setting policies and ways of thinking upon the rest of the people.
As an example, there has been a lot of talk about creating programs that push for more professors/executives from a certain group of people. I can see today that they just ended up pooling from the same middle-upper classes that has been filling universities for years. On the other hand, the people with no voice are still in the same situation. You have no idea of how harsh life is for lots of people.
There seem some sort of air going around that we are a better society compared to the past. People grab random numbers, like the number of people above some poverty line, to say that we know better.
Even if we might have more technology compared to humans that lived one thousands years ago, I doubt that we are wiser.
The are books that have been written and rules that have served humans well, as a society as a whole. There are mistakes that humans have repeatedly done in the past. Why did we forget?
One of the consequences of this kind of thinking in this law-less society is that not everybody is born equal. I am disgusted by the disregard of the ruling class that they know better.
A lot of this is implicit, but I can see this around having interacted with people from privileged and not privileged backgrounds (and everything else in the middle).
We need to come back to fundamental principles. And the first one, is the importance of traditional family values, that see people in the unit of the boat.
If you do not see the catastrophe that is upon us as humans, then I hope you wake up.

Cipolla 24 Jan 2025 21:11 UTC
3 points
0
in reply to: Nathan Helm-Burger’s comment on: Cipolla’s Shortform
Yes, I have tried :) But I still value expert human intuition.
I think at least 50% of the doctors will go away in the future, as they are just trained to give the usual stuff, and if they disappear tomorrow an LLM would still give similar results (actually, even a monkey with just two basic medicines would be a good enough doctor).

Cipolla 24 Jan 2025 13:01 UTC
3 points
0
on: Cipolla’s Shortform
Not a proper quick take and perhaps off-topic on this forum. But given that I know some people here are into health I give it a shot.

I would be very grateful if someone could point me to some excellent doctors around Europe, website, or some sort of diet that can be good to improve my health.
I have been having some health issues:
- getting sick frequently, feeling feverish most of the time
- Difficulty breathing, with some pain sometime when I am also sick
  - It turns out I have recently developed asthma too. Initially a couple of doctors thought it was anxiety, and after a year and a half I really pushed to get a proper test.
- A burning sensation on my left chest, since I took the covid vaccine, that gets worse when I get sick
- Slight hypertension
- And most importantly a continuous head confusion (similar to when I used to get a fever) since two years ago! Sometimes it has head tingling, with hands and feet tingling too, and sometimes a headache.
- I am fit, and try to workout when I can.^[1]
All the doctors that I have met do not have a single clue, and do not seem interested into solving the problem or investigate.^[2] I am not even looking for a cure now. Just a diagnosis.
I am not rich so my budget is limited. But life has become so difficult.
1. ^
  With my breathing problems I can do a 10km at around 4:40 km/min, but suddenly not on my top of my game (4:10 km/min).
2. ^
  If it is not an easy ibuprofen or similar, they quickly give up. I have a big suspicion that doctors are not trained well, so they might be all effectively incompetent at dealing with situations that are not solved by the usual medicines they prescribe. I plan one day to write a post, as methods of rationality and AI might help in diagnosing situations.

Cipolla 30 Dec 2024 15:34 UTC
3 points
0
on: Cipolla’s Shortform
There should be more talk about concentration of power, decoupling from the masses, and social fragmentation around here.

Disruptive technologies amplify the economic inequality gap within countries, and between countries.
- New technologies, like automation or farming, quickly benefit a few number of people, basically the owners
  - As new technology emerges, the general population will see wage stagnation, job displacements, and will be able to get some benefit from the new technology only after a long time
- New disruptive technologies tend to directly benefit fewer and fewer people, while a larger side of the population has less to offer
  - While it is true that often new technologies will increase the benefits for everybody, e.g. more food, on the other hand the gap between people is humongous (e.g. food quality is way different), and who owns the food can steer the choices of the many
- At some point, the few people will not need anymore the general population to reach certain goals, and this can go on and on, especially with an exponential increase in technology (just look for the fact that now tech companies basically dwarf most of the others, and in the past century we had manufacturing companies, so these has never reached the power of the tech companies)
- Coupled to the fact that human individuals in the most developed countries are becoming more and more individualistic, with capillary technology it will be very easy for powerful players with amazing technology to steer humanity
Which are the consequences of this? I do not see many people discussing it on the forum.

While it is true that the AI itself might doom humanity, I consider human dynamics to be a greater threat (just for the fact that investing in AI can increase the likelihood of these fewer people to own the world, and this might increase the likelihood of humanity AI doom).
This is a better defined problem to the usual AI safety talk. I think it should be something that needs to be addressed asap with public discourse. We need to focus on the right things in my opinion.

It might also have some implicit benefits on usual AI safety (and to be sincere, I do not see current AI research level at the level of research in math or fundamental physics, where first principles approaches and taking your own time to deeply think about problems are preferred, but this is another topic for another day).

Cipolla 5 Nov 2024 21:07 UTC
1 point
0
in reply to: brambleboy’s comment on: Matt Goldenberg’s Short Term Feed
Do you know why the error bars in the replication are smaller than the original one? (just more people?) And with which confidence is the null hypothesis (difference = 0) is rejected in both cases?

Cipolla 21 Oct 2024 7:57 UTC
8 points
3
on: Cipolla’s Shortform
I noticed the most successful people, in the sense of advancing their career and publishing papers, I meet at work have a certain belief in themselves. What is striking, no matter their age/career stage, it is like they are already taking certain their success and where to go in the future.
I also noticed this is something that people from non-working class backgrounds manage to do.
Second point. They are good at finishing projects and delivering results in time.
I noticed that this was somehow independent from how smart is someone.
While I am very good at single tasks, I have always struggled with long term academic performance. I know it is true for some other people too.
What kind of knowledge/mentality am I missing? Because I feel stuck.

Cipolla’s Shortform

Cipolla21 Oct 2024 7:57 UTC

1 point

15 comments1 min readLW link

[Question] How do you finish your tasks faster?

Cipolla21 Aug 2024 20:01 UTC

4 points

2 comments1 min readLW link

Cipolla 21 Aug 2024 12:37 UTC
1 point
1
in reply to: Jiro’s comment on: Caring only about your child may increase human suffering
This statement would not apply here, because:
- the post is about caring too much/only about something/someone we care about, and it does not imply about not caring/not favouring yourself
- caring about your child has a priority on caring about you, hence the topic of the post
Indeed, I would argue that it is important to care about oneself, if you want to help others for example, or have a nice life (or for example if you want to get a promotion, etc...).
But when you care too much, and do not look into the big picture of your actions, bad things can happen. Simple naive example: European lords/current dictators sucking up the money to build for themselves better castles/palaces instead of pouring some into society.

Cipolla

Look­ing for back­doors in Jane Street LLMs

Cipolla’s Shortform

[Question] How do you finish your tasks faster?

Looking for backdoors in Jane Street LLMs