lcnPylGDnU4H9OF

> (If something that took 10s now takes 1s, that's “ten times as fast”, not “zero point nine times faster”.)

It still works with speed as the measurement (not as "zero point nine" but just "nine"), you just have to inverse the relationship. Going from 10s to 1s can still be described as "9 times faster" but it's because, less intuitively as you say, "faster" is comparing the 1s to the 10s (as in, "9(1s) plus 1s is 10s") rather than the other way around ("9(10s) plus 10s is 1s" is obviously false). This works because we know that 1s is less time than 10s; "faster" inverses the numerical relationship for time because increasing it inherently means "slower". Indeed, the difference between them is 9x where x = 1. And 10 to 0.1 is 99 times faster because 99x where x = 0.1 is 9.9 (measured by throughput, assuming only the rate changed, it's also 99x more things and 100x as many things).

(Not that I think you don't know all this, just nitpicking, really.)

Compare this thread to one two years ago announcing Trump's Presidential victory. That wasn't flagged; thousands of comments; still US politics. No, it's clear partisanship that's hidden behind an implausible veneer of anti-politics rhetoric.

https://news.ycombinator.com/item?id=42057647

> stick to one topic: was the US the worst place for black people in the world? No it was not. Provide evidence that it was. Then I will point out to you that it was worse in other places.

Can you link the comment where anybody said it was the worst place? Brother, your claim is that, because some guy said that America is the best place in the world for Africans it must be true, and other people are just pointing out the things that make America bad for Africans. Nobody even said it was the worst!

Stop changing the subject.

> It's not a bad copy, or at least that's not what the person you responded to is insinuating.

FWIW, GP didn't write "a bad copy" but "bad copy", referring to the marketing sentence quoted by OP, rather than the model that was just released (as I presume "a bad copy" sounds, maybe like it's called that because it's [again, presumably] distilled?). "Copy", especially in a marketing context, is a written message.

> Note to others who are blaming Amthropic in various ways. This is what taking responsibility is.

It is way too early to tell if Anthropic is "taking responsibilty" for this. I'll be awaiting the news that they started aggressively hiring human support staff to field user complaints so they don't continue to do shit like this. That is taking responsibilty.

> The report is highly sensationalist in the "zero-click" aspect. It details to some degree the aspects of distribution and C&C but goes very vague the main point - what was the mechanism that activated that zero malware in a zero-click fashion.

It read to me like it was assumed to be using OS/browser/application-level exploits to get further access to target devices. I suspect the CitizenLab report is light on details in particular because they don't actually know them:

> In most cases, the user who sees the ad is not selected for targeting, and the server will not return any JavaScript to be executed.

> A large number of ads placed by Discover-Tech contain a script tag pointing back to what the company refers to as a “NOAS server” (also called a “getAd server”). This tag causes the ad, when rendered, to connect back to Discover-Tech, submit advertising identifiers for the user (either a cookie ID or a mobile ad ID), and automatically execute any JavaScript code returned. The NOAS server decides, apparently based on a combination of the source IP address and the advertising identifiers, whether the target is eligible to receive any JavaScript. If so, then code is returned. If not, a 0-byte response is returned. The NOAS server is a key part of Discover-Tech’s strategy, allowing Discover-Tech to avoid directly sharing malicious code, or even pointers to malicious code, with the companies who serve and display the ads.

Agreed it's against the guidelines as written but the interpersonal exchange is no different than someone quoting a blog post (and then not linking to it, to be fair). Something like this doesn't feel strictly bad, assuming disclosure and a link to the chat, sources included.

That article also has... let's say an unconventional writing style. Here's the first paragraph:

> Expected value. Expected value expected value. Ketamine, Kelly criterion, child support. Roko’s basilisk; Adderall; girlfriend. Shut up and multiply. Escorts. Expected value.

And the rest goes on like that, but with different in-joke buzzwords.

https://morewrong.lol/?post=against-taking-my-son-to-soccer-...

> Many, many, many Fortune 500 companies use Israeli cybersecurity services.

It's possible for this to be true, and that they spy on said companies on behalf of the government of Israel. It makes it easier to do so.

It's a bit unclear; the point the article makes explicit is that the commission raises prices when coupled with the most-favored nation agreements. There is the mention of "supplier-funded promotions" as "kickbacks", which might be from direct bookings.

You might enjoy this: https://en.wikipedia.org/wiki/Nominative_determinism. Not quite the same thing, more like if Plamen was a firefighter (or arsonist, I guess), but it's what your comment reminded me of.

I dunno about "nobody". I think of tulips, the Brooklyn Bridge, the Eiffel Tower, and Google Play gift cards to pay back Microsoft (after accidentally inputting too many zeroes in their banking software on their secure server).

https://en.wikipedia.org/wiki/Tulip_mania

https://en.wikipedia.org/wiki/George_C._Parker

https://en.wikipedia.org/wiki/Victor_Lustig

https://en.wikipedia.org/wiki/Overpayment_scam

> The CFTC is supposed to have 5 chairs, with no more than 3 in the same party.

Of course it's a policy that helps the two-party system become further entrenched. It should be no more than 2 from the same party. (Not that it matters given the fact only one is appointed, but still.)

Sometimes it is, sometimes it's not.

> It is also true that there is immense marketing value to "strawberries" being the first ingredient, and if "sugar" was listed first they would see a drop in sales.

To add to this, I recall an actual marketing campaign from one of the big US cereal manufacturers (I forget which) about the first listed ingredient in some of their cereals being whole grain, explicitly selling their product on the fact that the plurality ingredient of it is healthy.

Granted, that by itself doesn't mean much, but if they also split sugar and high-fructose corn syrup, their claim of "more whole grain than any other ingredient" could easily fall under technically-correct-but-misleading. Of course, a shopper can get more useful information from the nutritional facts but most people shop by the pretty picture on the front of the box (and the friendly message on their screen) rather than the technical label on the back.

It reads as an example of English writing that was once considered "proper". The point I see is that it shows how this complaint about AI writing is history rhyming with itself again.

> I think you picked the cutoff (70) so that the numbers look worse than they are.

Reminds me of a point I heard about sports statistics. If a given team is reported as having "won 4 out of their last 7 matches", there are good odds that they also won 4 out of their last 8 matches. Also pretty good odds that their seventh-last match was a win, otherwise the stat would be 4 out of 6. (It could also go the reverse direction if the goal is to make it look worse than it is.)

I see the point (though I'd also say they're arguing it poorly, and that it's a departure from the point in their initial comments; note my comment is also not an endorsement of it). OP said that the LLM vendors should face legal problems for making the image, but the person they supplied it to was not tricked by its fraudulent nature, therefore the LLM vendor did not commit the fraud. Ostensibly, that is the dishonor of the person who prompted the model to create the fraudulent work.

Recall also that you wrote this here:

> People have free will. Bad people know what right and wrong is, and they chose wrong.

The context being that the "bad" person is such simply because they chose to break the law. Even if it doesn't match your thoughts, surely you can see how one might read your comments as both claiming that laws are not morals, and also that those who break the law made the "wrong" (moral) choice (put another way: laws are morals).

Breaking the law is not strictly a moral failure. It can be, but they have an "orthogonal" relationship. To that point:

> I did not say there was no injustice happening. Nor did I say that all laws are just.

Choosing to follow an unjust law is a moral failure. That moral failure is baked into the word "unjust". This blanket statement that the choice to commit a crime is a choice to be "bad" is not self-evident (or it's laughably tautological) and, regardless, is contradicted by the assertion that not all laws are "just".

Of course, one can assume you have more nuanced thoughts than that; the point of my coment is that you've not communicated that nuance here, so you appear to be contradicting yourself: saying both that laws are morals and that they are not morals.

> We'll make up more games.

This reminded me of something. https://www.sbnation.com/a/17776-football

That could alternatively be seen as a reason we should make employment less hostile to workers and encourage less car-centric city designs.

No clue what OP's intention was but I'll note that the software on object recognition cameras typically makes use of a vision model.

Not that surveillance capitalism is needed by society, but that surveillance capitalism needs privacy violations in order to thrive.

I believe what parent commenter is suggesting is that it would be ridiculous for lawmakers to call it a crime to die, say without verifying you are of the legal age to do so. It would be like declaring that breathing is legal: why does that need a proclamation?

> Explain how a privacy that protects you and me will somehow not protect a pedophile.

I think that it will, and that's okay. The US Bill of Rights necessarily protects criminals as well as innocents. "The optimal amount of [crime] is non-zero." (https://www.bitsaboutmoney.com/archive/optimal-amount-of-fra...)

> For whatever reason, I can't reply to your comment further down.

In case you happen to read this, that is (presumably) due to an anti-flamewar feature on HN, which hides the reply button for the first few minutes of posts in deep comment threads. You can either wait to comment, or comment by clicking the "X minutes ago" text next to the commenter's username.

This is a perfect example of a psychological projection. Your comment describes itself.

I see what you're saying but it strikes me as naive to suggest that a felony and misdemeanor are different merely in magnitude. I suppose that was my mistake given my initial examples. In practical terms, one is a means of voter disenfranchisement, among many other life complications like "failing" employer and landlord background checks. One thing being a felony and the other being a misdemeanor can be seen as a difference in magnitude and as a difference in kind, and I was talking about the latter.

Then you don't see why stealing $10 is a misdemeanor and stealing $1000 is a felony (depending on jurisdiction, but whatever you get the point). Not saying you should see why, just that many clearly do for it to be public policy.

But it makes me wonder: do you think the $10 theft should be punished more heavily (minimum 1 year in prison), or the $1000 theft punished more lightly (maximum 1 year in prison)?

This "they did bad stuff so we have to do worse stuff" idea strips the people doing the second bad thing of their agency. That's false; they have agency, and can choose to do better.