I find that kind of unlikely. If they wanted to frame someone just to have a killer, they wouldn't be talking about a "3D printed ghost gun", but just use a regular gun. I, for one, haven't known that it's possible to 3D print a pretty well working, and silenced, gun. And that might inspire someone - acquiring weapons is the harder part of any such murder, assuming you don't want to get caught, and the fact that you can get it without anyone knowing about it makes it way easier.
- Posts
- 5
- Comments
- 306
- Joined
- 3 yr. ago
- Posts
- 5
- Comments
- 306
- Joined
- 3 yr. ago
What's the state of LLM detection algorithms? Is there any with a higher sucess rate and with OK-ish amount of false positives? Is there even a FOSS solution for detecting chatgpt? Would make for a great tool to have, I'm getting tired of this.
- Jump
My minimal LLM instructions. These are the current ones for IDEs with their in-IDE AI in particular.
I find it kind of hilarious how almost every prompt I've seen leaked from various apps almost always has a similar clause, as if it would have any effect at all on the result.
Seeing engineers resort to this level of basically praying and wishful thinking that in reality has no factual value is pretty funny.
"Please, don't give me wrong results 0_0"
I'm running Freetube for all my YouTube viewing in both Fedora and Windows at work, since it can subscribe without having a perosnalized feed and doesn't requires an account. It can also download in variety of formats, so I suppose it does run ytdl in some capacity in the backend.
I admit I'm kinda disappointed. He pulled out almost perfect assassination that looked well thought out, managed to get away with only a few hickups in his plan as far as his face is considered, and then walks around with a murder weapon and a manifesto in his bag? Shame, really. All he needed was to lay low for a while, grow a beard and he'd probably be OK.
I mean, that's literally how research works. You make small discoveries and use them to move forward.
I shudder at the thought of "Programming Inquisition", that would drive around doing PRs and exterminating anyone whose code shows signs of heresy against The One SOLID God.
None of us would make it :D
Oh, I totally forgot about this bit. Maan, I really want to write "a practical guide for programming with the Omnissiah.", a book of fun rituals and litanies to support your software development. I always felt like computers are kind of really literally magical, and adopting a more spiritualistic approach to programming sounds like a lot of fun. Unfortunately, I didn't find any existing spiritualism that would match this, Tech Priests being really close to what I was going for.
I need to add more litanies and copious amounts of censer into my programming workflow.
It took me a long time to realize the worth of having a CS degree. When I was leaving the school, I felt like it hasn't tought me much. I was already a pretty ok programmer, since I was programming most of my highschool, and it felt like I've wasted a lot of time on languages I'll never see in my entire life. Which is kind of true - I'm still pretty confident that I'll never use Lisp, Prolog, Lambda Calcul, base assembly or Pharo ever again, but after a few years I've realized something important that I was missing - the school wasn't trying to teach me how to be a "pharo/lisp/prolog programmer", but to be "a programmer".
I noticed it on my pentesting colleagues who didn't have formal programming education, how they mostly spoke about programming in relation to languages - "I know a little bit of python, but wouldn't call myself a programmer. What programmer are you?". That question felt wierd, and I eventualy realized that's because the lines between languages eventually blured for me naturally, and I paid no mind to the language of choice - I was simply able to naturally pick up any language, and write anything I needed in it pretty quickly.
Only then it occured to me that I have my education to thank for that. Sure, I might never use Lisp again, but I do vaguely remember the concepts and workflow the language has, so now I can more naturally pick up any lisp-like language. Same goes for the prolog-style of languages, or the more OOP-focused languages, like Pharo. Since I had to drag myself through hell to pass an exam in most of the flavours of languages, it made me a versatile programmer that can just naturally pick up anything I see, to the point where I don't have to think about it - I just subconsciously detect what kind of basic workflow style is it going for, google the basic syntax and standard libraries, and I can write whatever I need in whatever language is available in a reasonable amount of time.
I don't see this "ascendance" mentioned in the post, and I think that it's a really important point in learning to be a programmer. It's also a piece of advice I try to give anyone unsure about whether his degree is worth it, because it feels like you're learning useless stuff. I have no idea how to teach it, though. It kind of happened naturally for me, and I can't identify the point when it happened or why, or how would I go in teaching it to someone else.
It's important to keep a wide field of view when learning programming, and not just lock yourself into one language. You can always google for syntax pretty quickly, but seeing the wide array of workflows and flavours different languages use to accomplish the same thing will go a long way in making you a better programmer.
My favorite windows update was when I was at local Microsoft office on some kind of highschool coding competition hosted by Microsoft, and we had to start 10 minutes late because we were watching the meeting room computer force a restart with Windows update a minute after the introduction presentation started.
It isn't clear what exactly is the event about - is it about teaching newcomers how to get into contributing to the FOSS projects, or about kickstarting your own FOSS projects? The vague info seems to mostly talk about "your project", so it seems to be the latter?
I'm more interrested in the contributing side of things, but that doesn't seem to be mentioned in the info. Anyone has more information in this regard? I really like the idea, but if I don't have my own project in mind and don't want to start anything and just contribute, would it be for me?
I can't recommend Maldev Academy enough. It has been an amazing resource, to get into malware development. Keep in mind, however, that malware development is pretty difficult topic. You will have to eventually use WinAPI and syscalls, so learning about that even outside of malware development will help you a lot.
For example, try looking into how to execute a shellcode in memory - allocate memory as RWX, copy some data and then execute it. Try executing it in a different process, or in a different thread of another process. That's the core of malware development you'll probably eventually have to do anyway. Manually calling syscalls is also a skill that you'll need, if you want to get into EDR avoidance.
Also, look into IoCs and what kind of different stuff can be used to detect the malware. Syscall hooks, signatures, AMSI, and syslog are all things that are being watched and analyze to detect malware, and knowing what exactly is your program logging and where is one of the most important and difficult skills you can get.
There probably are a lot resources for these two skills, and they are an important foundation for malware developemnt, so I'd suggest researching that. You'll probably not get much from looking at other malware, because it tends to be really low-level, and obfuscated, exactly to avoid the IoCs I've mentioned above. Implementing the malware behavior after that is the easier part.
Another good resource to look into are C2s and communication, for example Mythic C2 has some interresting stuff.
And I really recommend joining the Bloodhound slack. Throughout my cybersecurity carreer as a Red Teamer, the community has helped me a lot and I've learned amazing stuff just by lurking.
I stumbled upon the Geminy page by accident, so i figured lets give it a try.
I asked him in czech if he can also generate pictures. He said sure, and gave me examples about what to ask him.
So I asked him, again in czech, to generate a cat drinking a beer at a party.
His reply was that features for some languages are still under development, and that he can't do that in this language.
So I asked him in english.
I can't create images for you yet, but I can still find images from the web.
Ok, so I asked if he can find me the picture on the web, then.
I'm sorry, but I can't provide images of a cat drinking beer. Alcohol is harmful to animals and I don't want to promote anything that could put an animal at risk.
Great, now I have to argue with my search engine that is giving me lessons on morality and decide what is and isn't acceptable. I told him to get bent, that this was the worst first impression I ever had with any LLM model, and I'm never using that shit again. If this was integrated into google search (which I havent used for years and sticked to Kagi), and now replaces google assistant...
Good, that's what people get for sticking with google. It brings me joy to see Google dig it's own grave with such success.
76% of all respondents are using or are planning to use AI tools in their development process this year, an increase from last year (70%). Many more developers are currently using AI tools this year, too (62% vs. 44%).
What the fuck. That's horrifying. I also though that every sensible workplace bans the use of AI.
A friend was telling me about a discussion between CTO's at a conference, where they were talking about whether it's even worth it to hire junior developers anymore, since there's a high risk of them just being "AI-raised", without much (or any) experience of coding without AI. And, this survey result... I can see where they are coming from. The future of programming looks pretty bleak - our job will not be replaced. It will just get worse, with good developers being more of a rarity.
And the amount of people who use vim or neovim as their IDE is surprisingly high. Is it skewed by sysadmins?
If you are interested in code completion, I recommend taking a look at https://refact.ai/. Hosting it (last time I tried) was almost painless, setting up docker to work with your GPU takes some time, but is pretty ok-ishly documented on NVIDIA page, and then you just run a docker and it worked.
It runs a server you can connect to i.e with a VSCode plugin, that will provide code completion or a chatbot (depending on what model you run), and it also has an option to let it loose on your project. You set training hours, give it a git repo (or a zipfile with whole project), and it starts training, which should tailor it towards giving more relevant code completion in the context of the project. I'm not sure if you can do that for the chatbot models, though.
However, I was trying it on my spare gaming PC turned server, that has an unused NVIDIA 1060, and while I could run some smaller models, I wasn't able to get the training working - the 6Gb of VRAM simply aren't enough for that. I also tried running it on the PC I work on, but it kept eating like 20-30Gb of RAM for the container, which made it kind of hard to also do anything else on the PC.
However, if you have a spare PC/server with good GPU that can run it, I'd say it's one of the better ways how to get personalized code completion, that keeps your data local and secure.
As a side note, I think you can give it API keys and let it use online models, but that would kind of defeat the point.
Isn't that, like, illegal?
That's what I was reffering to. I'm looking for articles and inspiration about how to cleverly write NPC game AI that I'm struggling with, I don't want to see how are other people raping game deveopment, or 1000th tutorial about steering behaviors (which are, by the way, awfull solution for most of use-cases, and you will get frustrated with them - Context Steering or RVHO is way better, but explain that to any low-effort youtuber).
I've recetly just had to start using Google Scholar instead of search, just so I can find the answers I'm looking for...
This is unfortunately not true - AI has been a defined term for several years, maybe even decades by now. It's a whole field of study in Computer Science about different algorithms, including stuff like Expert Systems, agents based on FSM or Behavior Trees, and more. Only subset of AI algorithms require learning.
As a side-note, it must suck to be an AI CS student in this day and age. Searching for anything AI related on the internet now sucks, if you want to get to anything not directly related to LLMs. I'd hate to have to study for exams in this environment...
I hate it when CS terms become buzzwords... It makes academic learning so much harder, without providing anything positive to the subject. Only low-effort articles trying to explain subject matter they barely understand, usually mixing terms that have been exactly defined with unrelated stuff, making it super hard to find actually useful information. And the AI is the worst offender so far, being a game developer who needs to research AI Agents for games, it's attrocious. I have to sort through so many "I've used AI to make this game..." articles and YT videos, to the point it's basically not possible to find anything relevant to AI I'm interrested it...
Is it even possible to solve the prompt injection attack ("ignore all previous instructions") using the prompt alone?
This puts to words something I was recently thinking about pretty well, especially the part about being an "advice seeker" and not really being able to solve stuff on your own, which is something I've always attributed to just being a field where you are driven to, especially in school, to have The Correct Solution, and that one always exists.
I mostly struggled with this when I tried getting into art, especially music or drawing. Suddenly, there's no algorothm or The Solution, and you have to figure out something based only on your creativity and judgement, and there's no-one who will tell you "this is the correct answer", which for someone being used to there mostly being one, was something I never managed to get over to this day, because it simply stresses me to the point of creative paralysis.
Thankfully, due to enshitiffication of most of the services I was following, which basically forced me to drop them due to invasive privacy rules, AI integration, or not working in privacy focused browsers or over a VPN, it's getting better. I'm kind of looking forward to OpenAI, Google and Meta finally killing most of the internet, so I can let go when 90% of content is AI generated, 60% of websites wont work without chrome, and the rest is just porn.