Відео

very nice Explanation

my ass

Welcome back! We've missed you and your super content. It’s not about the quantity but the quality, and you always deliver TOP info. Weiter so ;-)

thanks for the effort you put into this..

one of the best and most explained tutorials! Youre the best! I understood everything clearly

What are the implications for hardware? Put into a personal computer context, it seems we will be using local hardware, but mostly cloud compute. Because an orchestrator llm could break up any task into steps, or threads, they could easily use different hardware. Creating a PC with 12 low end cpus and shared Ram could work very well. The os AI would manage the threads. Different specialist ais could run different cpus.

why is the bash in leviathan 3 automatically in the hammoud solution?

I really like the way you explain every detail, maybe I am not that moron afterall, i managed to get to natas6 with you and it was fun 😎 Thanx

This is cool; nice job. Btw for shadowing it doesn’t need to be on test. You can run both on prod and just log the shadow

Really helpful!

Thanks for making this!

Wow! Thanks for making this

Thanks for making this video!

Finally got around to watching! This was super helpful and I appreciate you taking the time to break down these topics. I took notes and sent them to my work slack.

Impressive article and presentation thanks for sharing such a valuable knowledge

Hey, welcome back! Love you content!

Really useful video thanks for the summary, I've also given your video some explicit feedback 👍

Good video . More video on this topic

Curious to know what your background, if you are happy to share? As you are not a web developer but you have knowledge of Burpsuite and software, I take it. My background is software and I know some web stuff...

the fact that you are learning and succeeding and all this is recorded in this video is inspiration to all of us. Thanks.

thanks! congrats, subscribed + liked!

Very much appreciate your sharing. I found this share when questioning how to read audit reports, especially many audit reports. It gave me a huge guide.

Is there a way we can get your notes

Nice man. Im currently watching your videos on the ethernaut challenges.

Please are your notes public?

I love this series. Thank you for letting us learn from your experience and research. One thing I wanted to add for others. On level 9, instead of using ";cat ../../../etc/natas_webpass/natas10" you can instead just "cat" out the contents of the password file by using: ";cat /etc/natas_webpass/natas10" "/etc/natas_webpass/natas"X"" is the file location for all of the passwords for Natas. But understanding the concept he used in this video is certainly useful for scenarios where the file you are looking for isn't a given.

Awesome methodology and thought process into this..

Thank you for the video.

Thanks for the detailed explanation. The rest of Leviathans were a breeze (partly because I learned a lot from this one)

Apparently you still needed to import the Force.sol contract file and still create an instance of the Force contract at the same time, otherwise remix would have thrown a compiler error, atleast thats what it did when I tried to avoid them... All in all, appreciate it for the awesome content. You make it so seamless solving these challenges...

as all the serie, this video is a 10 ! thanks for your work and all the questions you're answering in 25 minutes. If i may, at the very end, you say that you become the owner of the contract "Delegate", are you sure about this ? You used the function pwn() in the context of Delegation, so you should become the owner of Delegation, not Delegate right ?

love when you say "that was probably not the greatest explanation", bro that was I-M-M-A-C-U-L-A-T-E

Its also good to mention that the latest solidity pragma versions 0.8.0 and greater now have an in- built underflow/ overflow checker that mitigates this types of vulnerabilities. So basically once it comes across an underflow or overflow vulnerability, it fails to execute the txn, unless you decide to disable the checker by using the 'unchecked' keyword, but why would you disable it right...

Hi, how are you doing in the audit? Have you completely disappeared?

Amazing content broo || Love from India

Thanks for your sum up (:

I think if we start designing OS's for AI, then it could complicate training data. I think if we design programs for humans, and then give AI the ability to use that, then it will be more stable. Ive been working on a cross system AI assistant out of GPT-4 for a few months now. Ive got it to where its an effective universal voice assistant of sorts. ive even added GPT vision, however It cant locate points, so clicking aint an issue.

I solved it without remix watching your previous video it was dead obvious what the solution should be

sounds too good to be true

❤

Recent developements from Deepmind in relation to LLMs memorizaing and regurgitating training data - not-just-memorization.github.io/extracting-training-data-from-chatgpt.html The risk still seems overblown, especially if coroporates are preventing their employees from using this tools. 🤔

But risk of leakage not in LLM itself, but in service - which saving your queries, responses, metadata of dialogues etc. You can provide some nda-info, which will be stored - that you can see it in chat with bot, and that's just simple db which stores your queries and responses. And it can be a risk with inside job in company which provide you these services

thank you for taking the time to put this together. it's clear that the productivity gains from using LLMs can be significant, but doesn't that value hinge on the dataset it references? i think organizations make LLMs most valuable by putting their data into the database it defers to. do you agree? this unfortunately increases the likelihood of the real risk you highlight.

The loom filler word filter is so distracting

This is a very good video. Thanks 🙏🏻

Very good. I learned about the existence of Vector Db's. w.r.t security, there should be a chatGPT paid option where a client certificate is required for the interaction to be further secured. I would have thought that any FinTech would have already thought about this, in order to leverage the productivity but to ensure next level security.

Tell that to the various Board of Directors. They're the technophobes not the rank and file programmers.

cache... they have your conversations bro

Why would it be ok to give openai my secrets in the first place?

I don't believe any chat data is used to train the model. There should be zero connection to the next model of GPT and the data inputs from any chats. The real danger is someone intercepting packets between the opening server and your enterprise servers if you are sharing sensitive information.