dns_snek

35 ms·

811.
by dns_snek 1y ago
The exact attack they described is less of an issue these days due to HSTS and preloading, but: - make sure you're connected to the expected official domain (though many companies are desensitizing us to this threat by using distinct d
812.
by dns_snek 1y ago
No ads but it's far worse than just annoying — for me. I get annoyed when a video buffers for 10 seconds due to a technical hiccup. Being made to wait for up to a minute with pretend-technical issues and mocking messages like "Why
813.
by dns_snek 1y ago
As a customer I'm angry that businesses get to use "hope and pray" as their primary data protection measure without being forced to disclose it. "Motivators" only work on people who value their job more than the dat
814.
by dns_snek 1y ago
Consider yourself lucky. Some of their A/B tests seem to be designed to psychologically torment you with videos "buffering" for 10-60 seconds before they start playing, navigation taking 15+ seconds. If that happens to you, t
815.
by dns_snek 1y ago
People don't "have" AI. People are being provided AI as a service by the corporations who invested billions into training it. If they ever make a leap where it can actually replace the workforce, it won't be provided for
816.
by dns_snek 1y ago
> Especially after SyncThing disappeared from Play Store (because it was basically a non-Android app behind a thin Android shell that couldn't easily be upgraded to more modern native APIs), there's nothing super-obvious here.
817.
by dns_snek 1y ago
Before HSTS you didn't need a valid certificate. When you typed "amazon.com" in the address bar your browser would first connect to the server unencrypted on port 80 which would then redirect you to the HTTPS address. If some
818.
by dns_snek 1y ago
Such is the nature of probabilistic systems. Generally speaking, LLMs read the top N search results on the topic in question and uncritically summarize them in their answer. Emphasis on uncritically , therefore the quality of LLM answers i
819.
by dns_snek 1y ago
In addition to everything that's already been mentioned, another obvious difference is that energy and water are finite resources that are already provided at relatively low margins. Cloud services are provided at obscene gross margi
820.
by dns_snek 1y ago
In practice this means that you get the same content farm answer dressed up as a trustworthy answer without even getting the opportunity to exercise better judgement. God help you if you rely on them for questions about branded products, th
821.
by dns_snek 1y ago
I've been trapped in a quasi-NDA on bug bounty platforms too. The vendor just refused to make the report public long after the vulnerability had been fixed, likely to cover it up in case of any resulting damages claims (it was a financ
822.
by dns_snek 1y ago
You're assuming that the choice is between immediate public disclosure and coordinated disclosure. Doing "the responsible thing" takes effort that is often disrespected (sometimes to the extreme). I'm so sick and tired o
823.
by dns_snek 1y ago
> parts of the world are being hidden from me There's a Black Mirror episode on that (and more), S2E4 White Christmas.
824.
by dns_snek 1y ago
The person I responded to? Yes I'm agreeing with them, just adding my own thoughts. Maybe I could've worded that better :)
825.
by dns_snek 1y ago
Fair use wasn't designed for AI, but AI doesn't change the motivations and goals behind copyright. We should be returning back to the roots - why do we have copyright in the first place, what were the goals and the intent behind i
826.
by dns_snek 1y ago
That's true but I'm pretty sure that the goal is to have a large number of individual sponsors. A handful of large corporate sponsors can later try to use their sponsorship to exert unwanted influence over the project.
827.
by dns_snek 1y ago
I didn't say it was common, I said it's possible because I've seen it with my own eyes. He's over 30, I've known him for over 15 years and we've occasionally collaborated on software projects. I have a really g
828.
by dns_snek 1y ago
I don't know why you would say that about Go, you're never more than one programming error away from creating a RCE vulnerability, no matter the language. Linked RCE should demonstrate that quite clearly, don't you think? Eit
829.
by dns_snek 1y ago
> If you enter into an agreement with someone Upon review, it looks like you don't have to enter into an agreement with anyone in order to use this software, except its authors under the terms of the AGPL. > exploitative behavior
830.
by dns_snek 1y ago
Yes but nobody is stopping you from adding your own proxy which enforces any type of authentication you like, and in my opinion that's the more sensible approach here anyway. I don't think it's sensible to expect every projec
831.
by dns_snek 1y ago
Decide what? Slapping a simple, naive login screen on top of a service that was never designed to fend off attacks from untrusted networks doesn't fix the actual issue, which is the fact that an administrator exercised bad judgement an
832.
by dns_snek 1y ago
Your use of LLMs is distinctly different than the use being described here (in a good way). You might ask "What do I need to pay attention to when designing this type of electronic circuit", the people at risk of cognitive decline
833.
by dns_snek 1y ago
That would not surprise me at all given what I've observed in a couple of people who outsource the thinking part to LLMs. One of them has dropped at least 20 IQ points and went from being able to grasp complex concepts with ease to n
834.
by dns_snek 1y ago
If you think that's the worst case scenario you're in no position to be making security-related decisions. That line of thinking hinges on a very dangerous assumption that Ollama doesn't have any critical security vulnerabili
835.
by dns_snek 1y ago
We don't know what Apple does or doesn't do. They might not be able to unlock iPhones in their normal state, but what stops Apple from instructing your device to silently install a specially crafted update which makes that possibl
836.
by dns_snek 1y ago
Yes I (try to) use them all the time. I regularly compare ChatGPT, Gemini, and Claude side by side, especially when I sniff something that smells like bullshit. I probably have ~10 chats from the past week with each one. I ask genuine quest
837.
by dns_snek 1y ago
Zig team didn't want to be beholden to the whims of outside sponsors which is an understandable position.
838.
by dns_snek 1y ago
> Gemini frequently just straight up lies to me I'm pretty sure they meant LLMs in general, not just ChatGPT. They all straight up lie to very similar degrees, no contest there. > The only people who are being homogenized or &quo
839.
by dns_snek 1y ago
You get an acceptable answer maybe about 60% of the time, assuming most of your questions are really simple. The other 40% of the time it's complete nonsense dressed up as a reasonable answer.
840.
by dns_snek 1y ago
All food (other than raw produce) is processed. Cooking meat over a fire makes it processed food. You'll have to be more specific.

More ›