Everything written in ancient Greek that is foundational to western literature, has already been translated, likely to a higher standard than most of the people trying to learn it.
Unless you wish to be part of an effort to further improve the quality of these translations, including to adjust them for the fact modern languages themselves are a moving target, just read those translations.
Modern Greek, on the other hand, is a living language with new art and culture coming from it. I may not be able to write "a cup of tea please" without misspelling tea, nor pronouncing it so badly they reply in English (as per my user profile), but this is infinitely more valuable than knowing if the ancient Greek character inviting people over for a meal is saying the people will eat the meal or be the meal.
Wow, what a way to write off something "foundational to western literature".
I studied koine Greek with my dad. Today, I study Aristotle alongside half-a-dozen English translations (the latest, Adam Beresford's Ethics, is hilarious, "like Han Solo and Chewbacca, Achilles and Patrocles" in the notes; his Aristotle uses "Perhaps...but that's a bit off-topic").
None of the English translations is as convincing as knowing the original vocabulary. Many phrases and idioms are still obscure or debated. Why should the student not want to look behind the curtain?
Finally, there is something bracing about knowing the ancient grammar. Greek has features long-vanished from English.
You would separate students into those who never need to bother looking a bit into "foundational to western literature" and those handful who are on a PhD track. Eventually, nobody would grow up to be recruited into the latter.
To be fair, there are nuances in the ancient Greek which are best brought out by some study of the language. The most frequently translated ancient Greek text(s) would be the New Testament, and even there you can see a lot of modern churches base their ideas on dubious translations.
I find ancient Greek not so helpful when it comes to etymologies. Some are helpful, but many are obscure or misleading. Climax comes from the word for a ladder apparently, and electron comes from the word for amber. There are stories behind both but they won't get you far. Any word beginning with psych- tends to relate to the mind, but the Greek means "soul".
> EU economy is not on par with the US economy. This is a dangerously old belief. That was maybe true in 2000 but not in 2026. EU GDP per capita is ~$48k and US GDP per capita is ~$94k. US economy is nearly twice as big. Quarter of a century of higher growth will do that.
I think per capita is not a useful measure here? The populations are unequal.
By nominal exchange rates, the US economy is estimated to be $31.856T this year; the EU's $23T; by purchasing power parity exchange rates, the EU is $30.678T.
Exchange rates matter for what actually gets traded, but they're also easily shifted by interest rate policies. But even with this, any simplification of economics sufficient to fit in a comment is going to be very misleading about questions of who is more or less dependent on global free trade, the US or the EU. Even the complexity you list: I suspect there's an office or five in various EU nations filled with economists trying to work out exactly what would go down if there was an EU-US trade war and how to remove the critical points of failure.
> US exports to EU are critical for the functioning of the economy (assuming you count tech services as exports).
> It would be catastrophic for the world if there was a serious trade war between US and EU but if it involved major disruptions to tech services the EU would fold within days.
Yes, but this is kinda the point of all the digital sovereignty stuff.
It was already weird to me, as an iPhone app developer in Germany making apps for Germans living in Germany where sometimes the only language option was German, that I had to tick a box while uploading apps confirming that any encryption in the app would be in compliance with US export laws*; now, it's unacceptable.
All of them can be racist, but are you sure it is useful (or independently, accurate) to say they're the same?
While Reform technically didn't exist at the time Farage had his "Breaking Point" poster and all it echoed from the mid 1900s, Farage was very much responsible for it and he literally owned Reform until very recently.
I do not think they are the same, but right now I think the Greens the worst, and Reform about the same as Labour. I also think labour attitudes have more influence on their policy and wider culture so are more dangerous, because they come wrapped in claims to be be anti-racist.
There are also smaller groups that are worse: Restore and the EDL.
My real problem is with people who equate Reform with racism which is dangerous for two reasons. It gives racists on the "left" a free pass, and it ignores smaller but nasty groups on the "right". My biggest concerns are the Greens (who are likely to win a lot of sears in the next parliament), antisemitic independents, and Restore Britain who won all the seats the stood for (albeit in the same area) in the recent local government elections.
Also an underlying attitude that it is OK to be racist about some particular group, or people in the wrong political parties.
From what I've heard, every LLM before Mythos (which you can't get, they'll call you if you're big enough) will have far too many false positives to be helpful, so I guess the best option would be to use an agent to help you (not lights-off vibe coding!*) take advantage of all the older tools like valgrind and closing all the compiler warnings?
* I presume I'm not the only one to find the agents tasked with adding unit tests will sometimes try to sneak through "open source code and apply regex to confirm presence or absence of specific string literal".
They can speed you up significantly, but you absolutely do need to pay attention to what they produce.
With all respect to the Anthropic folks, that's just marketing. (If they're reading this: let us into the program so I can be proven wrong here.)
I'm sure what they have is awesome, but it's clear that there are people out there with some decent prompts that are getting results out of widely available models as well.
The big thing we're sharing is: bulk scanning by random people in random geographies got a _lot_ better around January, it's widely distributed, and it's going to get a lot better regardless of whether that specific version of Mythos becomes widely available or not.
> prompts that are getting results out of widely available models as well.
Absolutely, and the "false-positive" issue people keep citing as why Mythos is so good is easily solved in the harness, simplest solution is starting fresh context with another prompt to evaluate if it's a false-positive or not, just adding that drastically cuts down the rate.
That is false. A year ago every LLM generated report was slop - more likely a false positive than correct. However in the past few months nearly every LLM generated report is real.
Everyone is going to see different results. Overall the general trend is AI is getting better. Although that might be partially people are shutting down their bounty programs which gives the incentive to generate slop.
This is not good reasoning. You're offloading your thinking to "Turso" for some reason.
You're also assuming that they haven't made the alternative judgement that instead of triaging the haystack of slop that they get in order to potentially pay out to someone, they should instead be spending that cash and effort on tokens to find bugs in their own codebase.
You should read the mentioned article. They have hired some of the people they paid out to, and some of those people were LLM-assisted.
The claim I'm rebutting is "in the past few months nearly every LLM generated report is real." If that were true, there would be no need to close the bounty. The bounty is to address approaches that they themselves may not have considered, so would still hold value if the claim held true, as outside individuals may still hold unique LLM-assisted approaches and perspectives.
While I think it's not coming any time soon*, Canada's current economic alignment with the US makes them sufficiently far from being a sensible EU candidate that the UK rejoining is still in many senses closer.
Not close, neither is in the foreseeable future, though the future is exceptionally foggy right now.
* at a minimum, I'd expect a reversal of Brexit needing both that polls show at least 2:1 in favour sustained for a year, and also that anti-EU parties like Reform weren't one of the top two polling parties
The US' ongoing hostility towards Canada is decimating the relationship. The auto sector chaos inflicted by the US has destroyed decades of cross-border integration that benefited both countries, and it is now resulting in the cancellation of major projects (like the Honda $15 billion dollar Ontario EV project being scrapped). The trust Canada had shared with the US as relationship that benefited both countries is gone.
The sad thing is that the relationship probably cannot be rebuilt anytime soon. It will take decades to restore trust that was damaged in a few short years. Canada is forced to diversify and build closer relationships with less volatile nations across the world. This is probably good for Canada in the long term.
Indeed. But EU membership? Even at speed and with mutual will on both sides, I would expect it to be the work of 7-10 years to get past all the hurdles to Canada becoming a full member of the EU. Even just in car manufacturing, while getting stability would help eventually, in the short term I would expect anything like this to add more chaos just by changing from US-centric to EU-centric rules; and then add agriculture, fishing, etc.
I think the one scenario that could work for Brejoin is if Labour rebranded itself as the rejoin party to stave off the threat from the Greens.
The Remain contingent is lost anyways. The other side of the electorate is turned off by their waffling. This would give them a cause to rally around and allow them to consolidate their electorate again.
With enough Brexit voters now either dead (they skewed old!) or having changed their mind, plus younger folks that weren't eligible to vote back then being very pro-EU, that might just do it.
There's lots of things where it's not particularly difficult for very rich groups, sometimes even very rich individuals, to be able to afford so much of a thing that by buying it they damage opportunities for everyone else.
This is a known limitation of the free market, and why almost nowhere is actually a perfectly free market, only more or less free.
> I'm pretty sure they are actively trained to avoid it.
I'm not sure who is doing what training exactly, but I can say that (inconsistently!) some of my attempts to get it to solve problems that have not yet actually been solved, e.g. the Collatz conjecture, have it saying it doesn't know how to solve the problem.
Other times it absolutely makes stuff up; fortunately for me, my personality includes actually testing what it says, so I didn't fall into the sycophantic honey trap and take it seriously when it agreed with my shower thoughts, and definitely didn't listen when it identified a close-up photo of some solanum nigrum growing next to my tomatoes as being also tomatoes.
> Besides, like, what would you do if you asked your $200/mo AI something and it blanked on you?
I'd rather it said "IDK" than made some stuff up. Them making stuff up is, as we have seen from various news stories about AI, dangerous.
"Well-unknown" questions are maybe the one situation where LLMs will say "I don't know", simply because of all the overwhelming statements in its training data referring to the question as unknown. It'd be interesting to see how LLMs would adapt to changing facts. Suppose the Collatz conjecture was proven this year, and the next the major models got retrained. Would they be able to reconcile all the new discussion with the previous data?
I grew up one town away from
this, which won awards for both best and worst architecture, the former from architects, the latter from everyone else: https://en.wikipedia.org/wiki/Tricorn_Centre
> Like the Nigerian Prince emails, its main purpose is to identify marks.
I may be overly-generous to the guy (bad habit, billionaires don't need or benefit from best-faith interpretations of the stuff they do, leads to sycophancy), but I think this may be more like grandiose delusion than a 419 scam.
The continual promises of full-self-driving, however, those definitely seem like a 419: up-front fees for promises never delivered on, repeated again with newer better hardware. What version is the hardware on now?
Unless you wish to be part of an effort to further improve the quality of these translations, including to adjust them for the fact modern languages themselves are a moving target, just read those translations.
Modern Greek, on the other hand, is a living language with new art and culture coming from it. I may not be able to write "a cup of tea please" without misspelling tea, nor pronouncing it so badly they reply in English (as per my user profile), but this is infinitely more valuable than knowing if the ancient Greek character inviting people over for a meal is saying the people will eat the meal or be the meal.
reply