• 1 Post
  • 1.79K Comments
Joined 3 years ago
cake
Cake day: June 20th, 2023

help-circle


  • Honestly though, the Russians do it this way specifically because it’s a deniable operation, and even if they’re caught, the criminal prosecution can be a nightmare.

    So, idk; maybe if they’re caught sabotaging critical infrastructure of sovereign nations that happen to be in international waters, and intentionally setting things up so that it’s infeasible or impossible to successfully prosecute… don’t bother with the prosecution, go a bit old-school, and let the crew and the vessel plead their case with Davey Jones. Yes, it’s admittedly an extremely harsh approach… but, you know, fuck around and find out. If normal channels are intentionally made impossible to pursue, abnormal channels become the only real recourse.

    Civilized societies need to understand that there are people and nations that will intentionally exploit the ambiguities that exist on the fringes of said civilizations, and sometimes, violent response is the optimal solution. And it doesn’t have to be “from here on out” - it just has to be done a few times in particularly egregious situations that are going to become a legal quagmire. It’s analogous to a bully who beats you up constantly, and then one day you snap and bring a monkey wrench in your backpack and break his kneecaps. He’s not gonna beat you up anymore after that.

    Edit: and, compared to the idiotic bullshit we’re doing in the Caribbean right now, I’d feel a lot less bad if we sent a couple sabotage-oriented Russian shadow fleet vessels to the ocean floor instead.




  • And the solution to that is to scrub the metadata and introduce some light content fuzzing (to avoid variable keyword/phrasing inclusion traps that can be used for narrowing down leak investigations).

    Like, yeah, there are hazards to doing an exfil dump like that, but if you know what you’re doing and understand the threat vectors that are likely going to be employed against you, you can cover your bases reasonably thoroughly - especially if you have a homelab with local ML/LLM capabilities that you can use (and, well, know how to use) to obscure/modify the precise phrasing of things such that it becomes way harder to attribute the leak source.

    And, I’m not saying “chuck it into your local model and cross your fingers” - it would be an element of the sanitization pipeline. If you are so inclined to do this sort of thing, you should absolutely do as much manual and deterministic verification and sanitizing that you can.

    It goes without saying that this is all at your own risk. But if you think it’s worth it, there are ways it can be done.














  • Or just realize that nobody fucking likes LLMs as much as the Captains of Industry want us to believe, and that the true power of this technical domain lies in more targeted and bespoke ML model generation and usage.

    ML is good and enables - and has enabled - some genuine generational leaps in science and technology. But LLMs are such a fucking waste of the technology’s potential. Not to mention, I’m extremely irritated that (largely due to Nvidia cornering the market) everyone is super gung-ho about a digital approach which amounts to brute-forcing neural nets digitally with shitloads of memory and highly-parallel compute, when it’s obvious to anyone with more than a passing familiarity with electrical engineering that an analog approach is going to be FAR more efficient in terms of resource and energy usage.