The audience for this sort of thing is currently quite limited. It may grow more in the future, especially if prices get out of control, but for now, it's a relatively small subset of people.
Yeah, when I read this sentence I thought, "well, maybe you're just incorrectly classifying non-obvious LLM writing as human writing, which makes your hypothesis unfalsifiable".
Seems like it's a cat and mouse game that won't end anytime soon. LLMs are constantly getting better at writing, so at some point people won't be able to tell reliably tell the difference anymore. I wouldn't laude Pangram so highly at this point, since we're not at the final round yet.
I'm not absolutely opposed to AI in principle. If it writes things worth reading, I'm not opposed to reading them.
The problem is when it writes things that have the appearance of making sense but actually don't, when it writes what could be better put in 10% of the words, when it writes with no coherent logic. I don't have the time or the mental energy to wade through that on the faint chance that there might be a gem in there somewhere. But I have the same objection to human writers who do the same things.
Write something worth reading. I don't have time for it otherwise.
I agree completely, but how does that relate to my comment (saying it's cat and mouse, the game may never end, and Pangram's current success isn't indicative of future value)?
It's saying I don't actually care. I don't care if Pangram can detect AI writing in the future. I don't care if it can't. I'm saying the question does not matter to me.
I don't care about not reading AI. I care about not reading slop.
But I guess I could care about one outcome of this game. If Pangram loses the ability to detect AI writing, but AI writing continues to be this nonsense that has no real point or logic to it, and so more garbage gets published that I want to not read, yeah, I guess I would care about that.
Please, it’s cost was $19B/year. We spend $1T/year on military spending, $1T/year on interest payments on $40T in debt, and $100B+ (so far!) on the Iran war.
Musk cut it through DOGE because he’s racist, not for cost efficiency.
He is after all at this moment talking on Twitter about how he regularly tells people who say they are Jewish that they aren't Jewish and if they don't have DNA tests to prove it he doesn't believe them.
Back to USAID: Musk, a prominent US defence contractor also gave as one of his reasons to cut USAID that it was a front organisation for the American Intellegence community.
So just before a big war he threw a curveball at long running intelligence operations. In his own, public, on-the-record comments. The same man who provides Grok AI to the Pentagon for use on classified info.
He also, when challenged about the deaths, claimed that he hadn't cut any funding for that stuff and that it was all still done by a different department so there was "zero deaths".
Which is a lie, but if you believe his lie, then a) he didn't save any money b) unless you count the stuff that was really the American Intellegence operations.
On the other hand if you don't believe his lie then all the dweebs online defending all the deaths as worth it to cut the deficit are defending a man who doesn't even bother to make that same argument.
Your point about interest payments on the debt doesn't seem to support the assertion that we should go into more debt. It's not like we have a choice about continuing to make interest payments, which makes it a false comparison.
Regardless, I'm pretty sure the definition of murder doesn't include defunding programs with the knowledge that doing so will lead some people to die. If that were the case, then Joe Biden "murdered" the Americans who were killed by the illegal immigrants whom he willingly let across the border (by not funding ICE or allowing them to do their job). It was obvious that by letting in millions of them, some of them would end up killing people. Would you say he is a murderer too?
Going to a war with Iran that cannot be won for Israel was a choice, yeah? There ya go, $100B in savings, no DOGE theatrics required.
It is apparent (hah) you have a rigid conservative mental model, you’re not going to change your mind based on the data. It’s fine. The data and facts stand regardless of talking points socialized.
(native US citizens commit crimes at a higher rate than immigrants, here legally or not)
Going to war with Iran seems like a dumb idea, but I don't remember Elon being involved in that.
It doesn't really matter whether native born Americans commit crimes at higher rates than illegal immigrants. We shouldn't have any illegal immigrants here, so every additional crime is one we should not have had to endure. Also, if you want to look at how crime breaks down demographically in America, there are some striking patterns that might not fit your preferred narrative.
This is the wildest part. It seems that sub-250MB apps are the rarity now. So every 4 apps I install is 1GB. I guess it's good for Apple, since they know you can't back up to your SSD so will then have to buy iCloud (unless you're savvy enough to open up Terminal and make a symbolic link to an external drive – the fact that they don't enable manual changing of the backup location is unjustifiable).
> AI subscriptions are too highly subsidized right now
I've been running into annoying limits with Claude recently. It gives me like 5 questions over the course of 15 mins and then tells me to wait 5 hours. When companies can change things up to make the base subscription nearly useless (the last question always gets messed up, too), then you realize the value of owning your own infrastructure.
You're right that $200/mo is much cheaper than comparable infrastructure. OTOH, you don't get to have a computer that can also be used for other applications, or which works when the internet is down. Also, you remain tethered to whatever pricing the AI companies want to charge. If AI pricing goes like Uber/Lyft did when the VC cash ran out, then we'll be paying much more in a few years' time. We could look back and think "I wish I'd bought my own setup back in 2026" if it's going to be inevitable.
And all this before you get into privacy/security/compliance stuff.
Well it's not just text to text. It's text to code, text to images, research. For me 200 is too much but my 20€ sub yeah I do find that worth it.
If I actually made money with it (I don't and I don't want to) the 200 one probably would start making sense. If I were using it all day I'm sure it would save me more than that 200 by simply being able to do more work.
Different people can interpret the same social phenomena in different ways and subjective experience can differ between individuals. That your experience differs from mine doesn't negate mine.
There are objectively far more hits when you search HN comments for "people" than "humans". So while I would not negate your experience, I would say that your assertion that "at least on HN, almost no one refers to "people" anymore". There are 5 million hits that would say otherwise, more than 3x as many for "humans".
Pretty cool. I wonder at what point it makes sense to actually try to guess what the actual word is and then figure out which words you should guess in order to avoid it.
