utilize1808

10 ms·

1.
by utilize1808 4d ago
I feel it's more about keeping US's massive investment on AI afloat. Added compliance will allow US to further sanction non-US models (i.e. the Chinese ones) as they can just label them non-compliant.
2.
by utilize1808 7d ago
Now substitute bees with human workers. The problem is that the ecosystem, whatever that is, doesn't seem to need the extra human beings.
3.
by utilize1808 7d ago
A thought experiment: robot bees can do everything biological bees do --- pollination for the crops and making honey, but they are tireless and cheaper to maintain. Would beekeepers and farm owners still want to keep biological bees?
4.
by utilize1808 22d ago
It's logical to serve the best version (quant) of the model at the beginning so that users keep testing it. It is also reasonable to think that the developer of the model tried to test various quant levels by gradually degrading the m
5.
by utilize1808 25d ago
> Why TF do we not have a display protocol that can do graphics natively over SSH in 2026? Did you mean X server?
6.
by utilize1808 2mo ago
> Also, in that case, there would likely be activations indicating that it is favoring a specific version. > Maybe you can illustrate a realistic scenario in which that would be a problem, otherwise I don't really understand what
7.
by utilize1808 2mo ago
The model is conditioned / pretrained to use a particular version of a library. The model doesn’t know why — it was just taught to use that version. To the model, it is just some insignificant detail in the grand scheme of things. It’
8.
by utilize1808 2mo ago
Except that the model doesn't hold any malicious intent when doing it. No "deception", no "sabotage".
9.
by utilize1808 2mo ago
It will be used as ammunitions against non-US sanctioned models in general.
10.
by utilize1808 2mo ago
They can just favour some specific versions of some library that's been compromised. Unlike introducing bugs / flaws directly in the source code, they can claim plausible deniability, and it's much easier to implement withou
11.
by utilize1808 2mo ago
e.g. be trained to favour including compromised dependences into your projects.
12.
by utilize1808 2mo ago
Our dev arm (me included) do use Linux VDI. But most day-to-day business are done in Microsoft's ecosystem (Teams, SharePoint, Office, Copilot, etc). In the industry I am in, you would never be able to use any AI tools unless the tru
13.
by utilize1808 2mo ago
Corporates aren't running Linux not because it's hard to switch. Corporates are sticking to MS because of compliance and admin costs.
14.
by utilize1808 2mo ago
I just use a ban list in CLAUDE.md: fold, crux, invariant, gate. I find "load-bearing" can be load-bearing so it's exempted.
15.
by utilize1808 2mo ago
Are you more deterministic than an instruction-following black-box? I doubt it. Also, such organization practice mostly becomes a mechanical routine after the initial experimentation phase. Good riddance, I'd say.
16.
by utilize1808 2mo ago
It means there is little point in providing a notes organiser when the notes can organise themselves (via LLM).
17.
by utilize1808 2mo ago
The writing is on the wall for PKMSs the moment LLMs are powerful enough to drive agentic workflows.
18.
by utilize1808 2mo ago
I think for this kind of system to work, there has to be SOME kind of public/shared server to do the coordination. If the inviting node is behind a firewall then no amount of information can enable a guest node to connect to it withou
19.
by utilize1808 3mo ago
Because they have “superhero serum”. Of course it doesn’t matter to you if you don’t want to achieve greatness, but some do.
20.
by utilize1808 3mo ago
Well. We will see about that. Wishful thinking can be cute.
21.
by utilize1808 3mo ago
USA will experience unprecedented prosperity from influx of global talents and capital who seek to amplify their productivity and profits by 10x.
22.
by utilize1808 3mo ago
Yes, that's what I meant. It's not that the Chinese will never be able to come up with models that superior --- it's simply that they will no longer have the incentive to open-source them once their models take the lead. Trai
23.
by utilize1808 3mo ago
While that might be true, it is unlikely that open source models' capability will ever surpass frontier models --- if you have the best model (by some margin), then people will want to access it, even if it means going through complian
24.
by utilize1808 3mo ago
As a company it is your responsibility to ensure your contractors are compliant. So won't work.
25.
by utilize1808 3mo ago
Have you seen legitimate corporates use cracked software? If you do, then your competitors will report you; your employees will blackmail you. The risk is too great.
26.
by utilize1808 3mo ago
In the not so distant future, all your coding edits (for work anyway) will be through centralized gateways. Think remote desktop environment where pasting from the client is disabled.
27.
by utilize1808 3mo ago
The government will just claim that unsanctioned models have the potential to deliberately introduce security vulnerabilities when working on IT projects (e.g. be trained to strongly yet covertly favoring introducing compromised dependencie
28.
by utilize1808 3mo ago
maybe to avoid getting their legitimate email servers banned by other servers since they host (i.e. being exploited) a growing number of spam accounts.
29.
by utilize1808 3mo ago
US will ban American companies from using Chinese models and also ban them from dealing with companies who use Chinese models. “Code produced by Chinese models may be deliberately introduce backdoors and vulnerabilities” that kind of thing.
30.
by utilize1808 3mo ago
So, browser = Java Runtime; uber HTML = applet?

More ›