HNHacker News
TopNewBestAskShowJobs

kadoban

7,571 karma · joined April 20, 2018

submissionscomments
kadoban··on OpenAI Says It Will Not Release Newest A.I. Model Over Safety Concerns
Did they fix it by not releasing a point release of an already existing model? No?

Did they just get a bunch of credulous "news" coverage out of it? Yes?

Will they just release this within the next weeks, at best? Yes?

Huh, funny that.

kadoban··on Newly Discovered Neo Asteroid 2026 SA8 Close but Safe Approach
> Is it possible that one day a dinosaur extinction size asteroid will be annouced on a collision course?

Yes, but it gets less likely as we keep searching more of the sky.

You can never _really_ say it's impossible, there can always be a cosmic bullseye heading our way from way off in interstellar space that we can't see until it gets closer.

But there is still room for quite a few rocks in the solar system to have our name on them too, last I checked.

kadoban··on Judge orders release of records on Nazi war criminals who settled in Canada
Germany chased the nazi war criminals out.

Did Russia, Japan or China chase out their war criminals to the same extent? I doubt it.

kadoban··on Show HN: Panda, the world's first personal AI computer
How did they nail down a price but no specs? Very odd.

They're targetting people who want AI and will spend thousands, but don't know enough to ask _any_ details? Why wouldn't that audience just be on chatgpt?

kadoban··on Google's first Suncatcher orbital data center test launches October 1
They use passive cooling _because it's in a goddamn vacuum_. Passive cooling is basically your only option.
kadoban··on Google's first Suncatcher orbital data center test launches October 1
I encourage Musk to launch all of his GPUs into space as soon as possible. Benefit away, creepy weirdo.
kadoban··on Google's first Suncatcher orbital data center test launches October 1
Look at what they're doing. It runs for 15 minutes and then sleeps for an unstated amount of time (I'd guess hours to days).

Literally nobody thought it would explode if you run it for even a nanosecond.

kadoban··on Google's first Suncatcher orbital data center test launches October 1
It's an offensively stupid idea. Anyone trying to sell you on it thinks you're stupid.

Look at the article: they threw 4 cards in a box and launched it. It works for 15 minutes and then it has to snooze for who-knows-how-long.

kadoban··on Google's first Suncatcher orbital data center test launches October 1
> The satellite, dubbed MVP, is about the size of a refrigerator. Inside, it has four of Google’s custom TPU AI accelerators, which are used to train AI models and run inference to generate tokens for AI workloads.

Yeah, this is a science-fair project. They threw some cards in a box and are paying someone to launch it to generate idiotic news articles.

kadoban··on ArXiv receives multiyear commitments to support it as an independent nonprofit
You think authors are the ones driving the $49 view fees? They are not.
kadoban··on Pentagon says overreliance on AI contributed to missile strike on Iran school
> The antipathy towards Tman rivals Musk and few other public personas and that, previously somewhat small crowd, appears to be growing.

That would make me optimistic as well if dictatorships governed with the consent of the populace. They do not.

kadoban··on OpenAI GPT–6 Astra breaks Enigma message that has resisted solution since 2005
My understanding is that there's a bunch of short, probably unremarkable ciphertexts out there that if anyone spent a good amount of time on (days/weeks), they'd break, but nobody really has the incentive to do so.

These models/agents are quite good at that type of thing, working through the tedious bits with available information and tools.

kadoban··on OpenAI GPT–6 Astra breaks Enigma message that has resisted solution since 2005
> i would assume yes, to throw off decyphering. even more impressive that they managed to crack it

Were operators of enigma machines aware enough of cryptology or the weaknesses of enigma, for that to have been done intentionally to prevent decryption? I doubt it, otherwise _many_ things should have been done _much_ differently by the operators.

kadoban··on Tinfield 1 is an open weight coding model from Nigeria that beats Opus 4.8
There is, IMO a lot of value in fine tuning existing models. If that gets them funded, great, who cares, right?

This is not their first release. BTL-2 (or -3 maybe? I forget) looked kind of interesting but it was annoying to run, needed a patch on llama.cpp, looks like they're improving their tooling.

I'm certainly interested in how this actually performs.

kadoban··on Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
Only because I hadn't previously thought of it xD Step up from the standard ass-hat for sure.
kadoban··on Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
How has software/driver support been? I got burned hard by AMD last generation or the one before. Things smoother now, or do you have to baby it like hell and pick and choose software that works?
kadoban··on Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
Oh, wow, they think it's just a smidge below the q4? That's crazy good if true.
kadoban··on Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
Honestly looks pretty good except whatever is going on with its booty. Is that an ass helmet? I cannot parse what's going on there.
kadoban··on Don't Just Say Hello
My problem with that is the answer of if I'm too busy depends on what the request/question is. Does it need a huge context switch? Is it something I like working on? Etc.
kadoban··on Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
You can run the ~4 bit quant(s) on 24gb, if you're not _too_ picky on context size.

This will hopefully be better, though it'd be a _very_ surprising increase in performace at the size they say. Would love to see more about how it benchmarks.

kadoban··on Original Sony PlayStation 2 security chip 'broken wide open' after 26 years
> I don't think acid deserves such a scary reputation.

Acid in general can be handled quite safely with the right mindset and equipment.

HF is _fucking_ terrifying though. Even many professional chemists (or maybe especially pros?) won't work with it.

It's far beyond just being dangerous to your skin and eyes. Bicarb and goggles sure won't cut it as safety procedures.

kadoban··on Fed hikes rates as inflation worries push up bond yields
Inflation is high, so interest rates need to go up to try to slow that, but the economy isn't doing amazing already, and higher interest rates won't help that.

Not to mention the US debt is _high_ as hell and bond yields mean that's more expensive.

And the country is run by a broken fool who has no interest or ability to fix any of that.

kadoban··on Suspected sabotage causes major Netherlands rail disruption
> Because the Kremlin wants europe to take an overt action to justify mobilization.

Do they? Why? They're already in a stalemate of a conflict, why would they want more fully-mobilized enemies?

kadoban··on We got admin access to Baseten's production GitHub
> It's not, in most juridictions at least

What did I miss they did that's illegal? It looked like it downloaded a public docker image, searched around inside, and verified that the key it found was still valid (without making any changes), and then immediately notified them about the issue.

kadoban··on We got admin access to Baseten's production GitHub
> I know I can’t try and break into my neighbors house even if I have no intent of going inside and stealing once I break the lock.

They didn't break in. They found a key that their neighbor dropped and returned it.

> Is this legal?

Generally, yes (though ask a lawyer if you're going to do security work). Security researchers do occasionally get legal flak though, depending on which idiot they annoy by pointing out issues.

kadoban··on Oil Executives Say the Great Fuel Crisis Is Here
> Braess's/Pigou's Paradox that more car lanes cause more traffic was found in the 1920's.

The paradox is that it _can_ cause worse traffic, not that it always does. American cities are built like shit for transport, but it's not as simple as "roads too big, therefore traffic". Sprawl is a much bigger (and more difficult to do ~anything about) issue.

kadoban··on Dario Amodei says China presents 'toughest dilemma' for his proposed AI slowdown
And does he have any reason at all to shape it towards slowing down?
kadoban··on LG says we're fake news [video]
You know you can monitor/reject things it attempts to do, right? For anything important I have it write scripts for what it wants to do, then I can review them for anything troubling.
kadoban··on Garry Tan wants US open-weight AI labs to 'distill' frontier models, too
Oh I'm aware that they're ~all assholes and many have gotten away with worse. It just seems like a poor argument that we're required to make it happen for them.
kadoban··on Dario Amodei says China presents 'toughest dilemma' for his proposed AI slowdown
So, the person with vast incentives to keep the race to better models going as hard as possible? Good luck.
← PreviousPage 3 of 34Next →