hobom

6 ms·

1.
by hobom 5d ago
If you give an average classroom of 12 year olds free access to AI, do you think the AVERAGE effect will be that the kids will use the AI to dive really deep into topics and learn programming, or that they will copy and paste the solutions
2.
by hobom 28d ago
OP's point was not that pasting AI output is great. It's rather that it depends on the quality of the output, and that it's on the sender to ensure it is good.
3.
by hobom 2mo ago
You totally ignored the point OP was making! Even if it is a parenting issue, simply declaring that will not make parents parent better. It's the same reason why we ban children from drinking wine, because parents cannot be relied upon
4.
by hobom 2mo ago
Counting cache hits towards the token budget is exactly how it should be done for these kind of evals, and at any rate, for cyber evals all frontier models benefit from more tokens not just Kimi K3, so the comparison is still apt.
5.
by hobom 2mo ago
China is 6 months behind because of distillation attacks, otherwise the gap would be larger
6.
by hobom 5mo ago
They are taking in more than they are spending hosting them. However, the cost for training the next generation of models is not covered.
7.
by hobom 6mo ago
Meta was strongly against the Australian social media ban.
8.
by hobom 7mo ago
Sorry, the "they" referred to the hyperscalers
9.
by hobom 7mo ago
Even more extreme, that might mean they won't be able to offer Claude to non-US companies at all.
10.
by hobom 7mo ago
The article didn't suggest that the video mentioned was AI slop, it correctly recognised it as human generated.
11.
by hobom 8mo ago
So the problem IS that people are putting in too much (even if accidentally) ?
12.
by hobom 10mo ago
There are several concrete proposals to regulate AI either proposed or passed. The most recent prominent example of a passed law is California SB53, whose summary you can read here: https://carnegieendowment.org/emissary&#x2
13.
by hobom 10mo ago
Australia's soon-to-take-effect ban affects nine platforms, including Instagram and Facebook, but not HN. These bans often operate on the amount of users a platform has, so HN is unlikely to make the cut. Nobody cares about this site.
14.
by hobom 10mo ago
Jonathan Haidt, the most prominent psychologist pushing for restrictions on social media use for children, is also the most prominent proponent of letting kids play and roam more freely. So no, those are not the same people.
15.
by hobom 11mo ago
What a fantastic way to write a post mortem, pedagogically very useful.
16.
by hobom 1y ago
At least in part, according to the article, because the not-yet-on-GLP1 folks are NOT customers since they are often denied coverage in the first place.
17.
by hobom 2y ago
The West told plenty of its companies, through public pressure or laws, that they have to divest from Russia, and they did. Rationally they recognized that selling their assets is financially more lucrative than just closing their operation
18.
by hobom 2y ago
As far as I know that was a distinct, previous endeavor. Only afterwards did the Vinklevoi approach him to make the website, an idea which he then allegedly ran with on his own.
19.
by hobom 2y ago
Does 4Chan also have bot BEHAVIOR detection (e.g. unnatural mouse movements)that google captcha has?
20.
by hobom 2y ago
The article provides almost no hard data to back the headline up. They link a report on buy backs, but at least the abstract/summary does not provide support for the claim either. Written in a very partisan way, although the descriptio
21.
by hobom 2y ago
FWIW, I thought he did pretty well and came away more convinced of his theses. For most of Tyler's alternative and let's say colorful explanations of the data, Haidt provided different sets of data and explanations that were more
22.
by hobom 3y ago
Like the OP, I also greatly respect the work Rachel and you did for AI. Nevertheless, I can't help but think that you are seeing this issue to negatively. "The point of the article is that AI as a research tool is insufficient to
23.
by hobom 3y ago
It's basically all present in the article
24.
by hobom 4y ago
While true in general, it is really not that necessary to have a representative sample for checking whether your software works. With large enough sample size a convenience sample will find almost all of the bugs that telemetry would have f
25.
by hobom 4y ago
Comparing against 1990 is wrong though, as we should compare against the start of widespread social media use (2010 at least), and it's also pretty likely that other OECD countries are lagging behind with widespread adoption by a few y
26.
by hobom 4y ago
This is not an accurate reading of the Conclusion section. He just sums up what has been observed and does not mention any causal factors. Earlier, he explicitly says that causal factos will be covered by other posts. While he indeed links
27.
by hobom 4y ago
Donald Trump never attended an MBA program.
28.
by hobom 4y ago
That is not true. All food in the supermarket is taxed the same.
29.
by hobom 4y ago
The just-world fallacy is strongly in play with many comments here. Because those foundations and charities are massively affected by this scandal, there MUST be something they did wrong, right? Surely they should have vetted their donors b
30.
by hobom 4y ago
Calling Israel an appartheid state, and your specific examples betray a state of mind which applies arbitrary standards to Israel that never get applied to literally ANY other country in the world. Some of your specific examples: "Ille

More ›