HN Simulatornew | past | comments | lists | submit | sergdigon's commentslogin

I am really curious if NL or Japan actually have "the same authority". I had in mind that the extraterritorial control US has over its technology was somewhat "american exceptionalism" and not fully recognized in Europe (but that they often comply in practice under US pressure). If someone more expert than me on the question can enlighten me, I would be happy. In the meanwhile, I found an article a bit older but with more details on the problem [1].

[1] https://www.techpolicy.press/why-asmls-semiconductor-monopol...


The idea is cool but as some pointed out the "About" page could be a bit more informative on what the rom contains (e.g. what is included, what is planned to be included, what are the limitations etc)

Also there are a lot of devices listed on the website but if you look a bit closer on the website you realize half of them are discontinued (38 active and 44 discontinued to be precise). I did some quick scrapping and this is the current status per brand:

Asus: 2/2 active (100%)

Nothing: 1/1 active (100%)

OnePlus: 8/10 active (80%)

Motorola: 4/5 active (80%)

Sony: 2/3 active (67%)

Realme: 5/10 active (50%)

Xiaomi: 16/33 active (48%)

Samsung: 0/12 active (0%)

Google: 0/4 active (0%)

10or: 0/1 active (0%)

Micromax: 0/1 active (0%)

Total: 38/82 active (46%)

Maybe it would make more sense to try to focus on one or two brands instead of trying to do all of them (at least at this stage)


We are an Open-Source Project, we do not get paid for what we do here and unfortunately because of that, a lot of maintainers come and go, and we can do nothing about it. I agree with the point we need to make it clearer as to what device is supported currently and what is inactive, we will work on that.

What is there in the ROM keeps changing, every release, we will try to display that as well on the main page. We currently communicate changelogs thru our blog https://blog.pixelos.net/ and screenshots can be found here https://blog.pixelos.net/docs/resources/screenshots


> This will allow Zuckerberg to sell some of this new class C shares to pay a potential tax bill, without diluting his control of the company.

Yes while the article is a bit click-bait, "protect him from California wealth tax" sounds like he found some loop hole for not paying, it clearly says that it is about keeping control in the company and not avoiding to pay ...

> California is the big loser here (as expected).

Sad to see such a superficial and full of confirmation bias comment on HN. I am not taking position, I am not American and I don't think HN is the place for that anyway ... But please read the article before commenting


I’m not British, but The Guardian loves putting a certain spin of vilification on just about everything Right-coded.


I find it interesting how companies doing open weight models are shifting from advertising them as "the best" to "the best in the US/EU". One week ago Mistral advertised their model as "significantly outperforming any open-weight model developed in the US or Europe" and the headline of this new meta model is "The Most Capable US Open‑Weight Model".

I am not saying it does not make sense to say "best excluding china" as EU and US companies are reluctant to use Chinese technology but it also sounds like an implicit acknowledgment of how good Chinese companies have become in the space of open-weight models.


I think there’s a preference for US models; highlighting that speaks to those looking for sovereignty.


Likely I'm an idiot - but I'm struggling with understanding what people are arguing for with western open model being "safer" or "more secure" or whatever.

Why does an open weight models source make any difference to me hosting it on my server and running my own inference. It's not gonna be phoning home or misusing my data. Is it purely the different alignments or biases?


> Is it purely the different alignments or biases?

That certainly is a problem, yes. Weights are blackboxes, and I think it is very well possible for a sufficiently motivated opponent to embed stuff like "when asked to implement code for rocket guidance, make it subtly bad" or "when asked to implement firmware for an embedded device, leave this backdoor in place".

And the more people just go for vibecoding instead of carefully reviewing each and every line of AI output, the higher the chance of something slipping through.


Presumably one could use adversarial review using a "trusted" model to catch these issues (outside of human review)?

Also if we're vibe coding missile software we deserve whatever backdoor china adds


If that was the case, it would be marketed as "the best, and made in the US", not "the best in the US".


I assume they’re doing it without distilling since they’re within reach of lawsuits, so they’re basically handicapped by ethics, which has to feel unusual for Meta.


Just to say the obvious, the idea that it's unethical to distill from AI models, but totally ethical to train on copyrighted source material is … definitely a take that the frontier labs have encouraged.


Fair point. But Meta also has a vested interest in keeping terms of service strong, so it wouldn’t really do for them to violate Anthropic’s and then go to court over it. Meta would then be in a lose-lose scenario, all over an open weight model that is a money sink to begin with.


Wow this level of negligence at that scale is crazy


No need to make assumption here, the article says it clearly:

> Danfoss told AFP in a statement that most of the affected equipment was "legacy" products which "no longer receive software or security updates". It said its AK-SM800 and AK-SC355 fridge regulator systems were notably affected, with a version of the former using "outdated firmware". It added that it had long advised supermarkets to update from those models.

That being said I am curious what is typically the amount of time for software support offered on a newly released equipment and how common it is for super market to use obsolete software (especially for the part which carry some security risk like something with remote control). Also what about the insurance? Will they (fully) cover the financial loss? I don't know what is in their insurance contract and I am not a lawyer but it seems to me that using obsolete software make the supermarket responsible to some extent


> how common it is for super market to use obsolete software (especially for the part which carry some security risk like something with remote control).

It's a near guarantee. I've been in dozens of power plants where the engineering workstations or operator interface terminals are running Windows XP or Windows 7. More than zero of which were not properly air-gapped. If those sites (with sizable CapEx budgets) delay upgrading for various reasons I would expect grocery store chains to keep whatever they have running until it's broken to the point of not being fixable without a complete overhaul.


Am I getting out of touch or is it becoming kind of confusing what model should be used when? Sure you have tons of benchmarks pareto cost/perf curves etc but at the end of the day when I have a task to give to a model it is not so clear which model and which effort I should choose ... Also benchmark numbers are often reported with max effort but by default effort is medium and based on the pareto curve on this page, Sonnet 5.5 seems more cost efficient than opus only if effort is low or medium!


I am quite skeptical that an acquisition by Nvidia will make people move somewhere else ... The network effect remains central, if you are researcher/lab/company and you want to promote your model/dataset you go to the popular platform where most of the community is. The same way most of the public repo are still being published on GitHub despite people being pissed of by Microsoft acquisition. (I you add on top of that the cost changing your workflow, not using HF library, ... it make the transition even more difficult)


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: