TristanDaCunha

Do you have an example prompt I can try where frontier LLMs will stumble on physics?

I cannot find a strong source to prove this, but it seems to have been published on this website in 2005.

Will this be helpful to terrorist groups, as they try to get current and future open-weights LLMs to help them create better and more devastating weapons of all kinds?

Which statement in this article applies only to Python?

Should stop and help with neovim

Same result on Firefox mobile

This whole discussion on caching and abstraction was completely befuddling to me.

> The vast majority of the criminal activity has been on the "counter-protesters" who actually showed up.

Your source supports the notion that the counter-protests were better attended, but I can't see how it supports your assertion that they outdid the protests for criminal activity.

This reply doesn't help me at all

You have it totally backwards. It's a much bigger catastrophe if we over-focus on "safety" as avoiding sexism and so on, and then everyone dies.

> AI can only be as destructive as the users let it.

Not really, I suppose you aren't familiar with AI alignment.

Many of the signatories aren't associated with any corporation.

Well this is a more complex case which includes combining negation with quotation and punctuation marks. I'll bet those factors explain why it didn't work.

> Using @cached_property feels like a bad code smell and it’s a controversial design decision

Why?

What methodology do you envision for this project?

Well I'm not willing to provide any proof of my identity, so you have no reason to take what I say as legitimate. Indeed, I've seen lengthy posts of utter nonsense from others who claimed to work at CA.

Usually, even the generally available facts are a lot better than most people's understandings of the situation. For example, most people seem to think that this scandal proved that Facebook sell their user data, whereas it actually is alleged that CA took advantage of the well-documented public API that Facebook used to have, where any app vendor could download data on friends of friends who had given permissions to the app.

If you have specific questions, I would be happy to point you in the right direction.

Disclosure: I worked at Cambridge Analytica.

Cambridge Analytica were accused of taking data mainly on what pages users 'Liked' - not private messages or news feed. (Unless you have a source?)

Obama campaign boasted about having access to the entire friends network. In other words, they know who each person was and what the connections between them were. Facebook openly said that they helped the Obama campaign get and utilise 100% of the friends network data on Facebook.

> Google "European growth stocks".

This short sentence contains irony.

I was looking for a second opinion, and found the comment by Glenn in that article was a good counterpoint. I won't repeat it here, but another point that comes to mind is that one may use a very small amount of milk-substitute in your hot drink, compared to drinking a much larger amount of coke. So it's not perfectly comparable, even if they were equally bad.

Highly counterinuitive.

I gratefully accept the information of your post, but I don't think it manages to make the intended point. Notice how many times you used the word "should", which was never a consideration in my post.

We could talk about should. As a starting point, it would seem to me most useful to preserve meaningful distinctions which convey information, and otherwise make language as fluid and relaxed as possible. But I'm not expert.

So weird -- when did we all decide to start saying "I was laying in bed" instead of "I was lying in bed"? I have noticed this shift over the past few years.

Laying has always been past tense "That day I lay in bed for hours" or transitive "I lay the books down". Chickens also could be laying in bed, if eggs are coming out of them. Lying has always been intransitive "I was lying in bed". But nowadays, "I was laying in bed" is almost all I see.

> Slavery was long gone and finished when master/slave began being used in computer science.

I don't disagree with your conclusion, but this is such an America-centric statement. Slavery is alive and well today. There are more slaves now than at any past time.

I think there might be a mistake in this article, or at least something requiring clarification.

> For convenience, suppose the binary tree is complete and has height H=2^K.

What is K? It's never stated. I'd usually assume H = log N, if N is the number of nodes in a balanced tree.

In pneumonia, areas of the lungs fill with fluid. This prevents the area from filling with air, which therefore prevents the exchange of CO2 and O2 between the blood and air.

I will assume that 'absorption' in this sense refers to the fluid. It's a good thing when this happens, because that that area of the lung can be accessed by air again.

Yeah, the Guardian was actually interested in the story because they were mislead by the statements of the CEO of CA and by Christopher Wylie into believing that CA was involved in a campaign for Brexit in 2016. They wanted to delegitimise the outcome of the referendum.

Funnily enough, the newspaper doesn't mention that Wylie didn't work at CA in 2016 - so he can't be considered a first hand witness or a whistleblower concerning the American presidential election or Brexit. He left in 2014 after obtaining the Facebook dataset to start his own political data consultancy, Eunoia Technologies. The company was based on the concept of micro-targeting(, which he falsely believed was a novel idea). He competed with CA for contracts unsuccessfully, and Eunoia was ultimately sued by CA for stealing clients. (This might explain why he was so willing to work with the Guardian to bring the company down.)

All of the information in the above paragraph can be found on Wikipedia, except for the parts in brackets.

You can read the report from QC Julian Malins or the ICO about the whole situation, they're found no evidence of CA retaining the data. But you can't prove a negative.

>FB investigated CA to make sure they deleted the data, in turn out CA lied and hid copies of the data.

That's not true - they had deleted the data as requested. I worked there.

If you make a lot of money, almost certainly the best thing you can do is earning to give¹. Your impact would be far greater again if you were to donate towards any cause other than one which mostly victimises wealthy people.² Instead of cancer victims, you could help poverty victims or animals in factory farms.

¹https://80000hours.org/articles/earning-to-give/#should-you-...

²https://www.effectivealtruism.org/articles/introduction-to-e...

Okay, thanks for the clarification.

I do understand why the article hasn't touched upon this, since it's just an attempt to diagnose the pattern seen in the overall male subpopulation, which is not seen in the overall female subpopulation. But I agree it could be interesting to look there too.

I think the article takes as a given that men's labour participation is tanking, while women's isn't. For most of the audience, this will be common knowledge. Here is a graph about female participation: https://data.worldbank.org/indicator/SL.TLF.CACT.FE.NE.ZS?lo...