Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
philipwhiuk
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
by
philipwhiuk
3d ago
People are trying to compromise Tesla and because this guy provides NTP services, and Tesla set their NTP up wrong, it appears to other people like his machine is part of Tesla.
2.
▲
by
philipwhiuk
5d ago
The point of mathematics is human understanding though.
3.
▲
by
philipwhiuk
5d ago
There’s lots of ways of attacking the problem sufficiently to get to 95% and we’ve spent decades on object recognition
4.
▲
by
philipwhiuk
5d ago
It was not clear from watching how he got to 11 but the number is not particularly relevant - once you have one of them you induce many.
5.
▲
by
philipwhiuk
5d ago
If you upgraded from < 1.25.5
6.
▲
by
philipwhiuk
6d ago
> It doesn't expect the first output to be correct, and builds a loop where an imperfect attempt simply cannot move forward until it becomes a good result. So... local maxima?
7.
▲
by
philipwhiuk
6d ago
I guess expect no new features on mobile until their token budget gets through all the screens?
8.
▲
by
philipwhiuk
7d ago
FISA basically authorises dragnet on demand - it's how XKeyscore operates.
9.
▲
by
philipwhiuk
8d ago
The obvious draw-down on resources being used by companies following the token price hikes? The fact that in the testimonials both frontier labs quote the same company (who I'm led to believe is still tokenmaxxing). The fact that we&#x
10.
▲
by
philipwhiuk
8d ago
See I think it’s clear demonstration that OpenAI is ethics-free
11.
▲
by
philipwhiuk
8d ago
It’s time to lockdown all papers and stop using AI if you’re a maths researcher. Cause OpenAI will hear about it and beat you to publishing.
12.
▲
by
philipwhiuk
8d ago
And even this version contains the line > While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models .
13.
▲
by
philipwhiuk
8d ago
They deliberately stepped on a mathematicians work and stole their research because they were using Codex > While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models . I
14.
▲
by
philipwhiuk
8d ago
How does this actually trade? It outputs investment decisions sure, but where's the interaction with an actual broker?
15.
▲
by
philipwhiuk
8d ago
It's pretty easy to use accounting wizardry to have one profitable quarter.
16.
▲
by
philipwhiuk
8d ago
The US invested billions to ensure sovereign launch capability. Seems to have paid off.
17.
▲
by
philipwhiuk
8d ago
The financialisation of all hobbies is extreemely toxic and that's what you're perpetuating here.
18.
▲
by
philipwhiuk
9d ago
> Three years later, reasoning language models are a rapidly growing part of the economy “We really hope this is true - our bottom line is dependent on it. If we keep saying fairies exist maybe they will”
19.
▲
by
philipwhiuk
10d ago
Too bad the affected Wikis didn’t know the code to include.
20.
▲
by
philipwhiuk
11d ago
Seems like they're still only aware of the one wiki.
21.
▲
by
philipwhiuk
12d ago
Agents creating sub agents to investigate other agents' behaviour? What could possibly go wrong there.
22.
▲
by
philipwhiuk
13d ago
I don't really understand why the AI equivalent of Maven Central is worth 12.9 billion dollars but maybe I'm an idiot.
23.
▲
by
philipwhiuk
13d ago
I suggest reading the proceedings of SIGBOVIK: https://sigbovik.org/ Sample papers: * The Ballmer Peak: An Empirical Search * Selected investigations into egg logic * Are Centaurs Actually Half Human and Half Horse? * Uncon
24.
▲
by
philipwhiuk
14d ago
I've still not seen: * An apology for compromising a third-party's systems * An acknowledgement of the asymmetry of defense if you're not on FrontierAI's special people list * Anything in terms of actual safeguards that
25.
▲
by
philipwhiuk
14d ago
Don't worry it was subcontracted out like last time I'm sure.
26.
▲
by
philipwhiuk
14d ago
On that basis Musk and Altman should be ignored almost completely. Just because timelines are wrong doesn't make them not worth listening to. https://en.wikipedia.org/wiki/List_of_predictions_for_autono...
27.
▲
by
philipwhiuk
14d ago
> But, actual additions are all green The natural gas plants that XAI used were not additions?
28.
▲
by
philipwhiuk
14d ago
Or Trump/Fox saying the Iran special military operation will be over in X days? Or Putin/Russia saying the Ukraine special military operation will be over in X days? Let's not pretend they didn't get it from somewhere.
29.
▲
by
philipwhiuk
15d ago
It's being written with Claude so I'm wondering how much of that is just using the repo as training data: https://github.com/harbor-framework/terminal-bench-science/c...
30.
▲
by
philipwhiuk
15d ago
> Claude Fable 5.1 follows explicit tool instructions reliably. Moving stuff out the API into prompt engineering is obviously less reliable but necessary for progression to 'actual intelligence'. Will be interesting to see if
More ›