21–30 of 60 posts
I get that it's a way to find out what GPT4 is capable of, but IMHO that defeats the point of a game like CTF. It's like playing with an aimbot. You may be beating the other players, but where's the fun?
[flagged]
Neat! My mind-blown moment with GPT-4 was realizing that it will often be able to tell you the output of the (unique, not available in training data) scripts it writes for you.
That is very interesting, which I'd like to try once GPT4 is widely available. I did try GPT3 with a CTF a few weeks ago, and it gave seemingly plausible code but outright incorrect answers.
// returns 1337
return result;
I get that it's a way to find out what GPT4 is capable of, but IMHO that defeats the point of a game like CTF. It's like playing with an aimbot. You may be beating the other players, but where's the fun?
If I, as a person running a CTF, did not want my players to do this, I would set up a few problems which would have incorrect (but not obviously so) "solutions" generated when fed to LLM.
The Shamir's Secret Sharing reminds me of the time I was playing DEF CON CTF Quals, and had the bright idea to try to scan for challenges. I found one - it involved hiding fragments of a split secret in a modified version of ADVENT. I solved it. Even when the board was fully opened, it was nowhere to be seen. You does your hacks and you takes your chances...
Neat! My mind-blown moment with GPT-4 was realizing that it will often be able to tell you the output of the (unique, not available in training data) scripts it writes for you.
It’ll tell you what a typical output for the command might be, and the more complex the script, the more wrong and full of hallucinations it will be.
There’s a huge difference.
Specifically, you have no way of knowing the difference between accurate outputs and inaccurate outputs, without running the command yourself, making it largely worthless.
Without access to environment, it’s not possible to magically know what the output of a command will be, it doesn’t have an embedded understanding of code, specifically not iterations and mapped data or mathematical functions.
For trivial obvious outputs it’s good, but it’s not executing the code; it’s generating what seems like plausible output; and if the output is trivially derived from the input, it’ll be impressively accurate.
…but, as the complexity of the task increases or the task deviates from “standard problem space” the triviality of generating accurate output decreases and it stops generating impressive outputs.
Tldr; yes, but it doesn’t scale well beyond trivial outputs.
Neat! My mind-blown moment with GPT-4 was realizing that it will often be able to tell you the output of the (unique, not available in training data) scripts it writes for you.
It is impressive, but no, it won’t. It’ll tell you what a typical output for the command might be, and the more complex the script, the more wrong and full of hallucinations it will be. There’s a huge difference. Specifically, you have no way of knowing the difference between accurate outputs and inaccurate outputs, without running the command yourself, making it largely worthless. Without access to environment, it’s…
The former is necessarily the case given the Halting Problem; the latter is falsified by the fact we can reason about code despite the Halting Problem.
Earlier quoted context omitted.
It is impressive, but no, it won’t. It’ll tell you what a typical output for the command might be, and the more complex the script, the more wrong and full of hallucinations it will be. There’s a huge difference. Specifically, you have no way of knowing the difference between accurate outputs and inaccurate outputs, without running the command yourself, making it largely worthless. Without access to environment, it’s…
> Specifically, you have no way of knowing the difference between accurate outputs and inaccurate outputs, without running the command yourself, making it largely worthless. The former is necessarily the case given the Halting Problem; the latter is falsified by the fact we can reason about code despite the Halting Problem.
> the latter is falsified by the fact we can reason about code despite the Halting Problem
i think wokwokwok's point holds true in practice.
Our patience and working-memory is far more limited than what is essential to accurately model all the necessary details of even moderately complex algorithms in our head.
One of the main reasons to limit code-complexity to improve readability/maintainability.
I get that it's a way to find out what GPT4 is capable of, but IMHO that defeats the point of a game like CTF. It's like playing with an aimbot. You may be beating the other players, but where's the fun?
Sorry to the author and thanks for the article, I learned about lagrange interpolation.
Earlier quoted context omitted.
That is very interesting, which I'd like to try once GPT4 is widely available. I did try GPT3 with a CTF a few weeks ago, and it gave seemingly plausible code but outright incorrect answers.
I did some advent of code exercises with GPT-3 and it often ended the script with something like // returns 1337 return result; Sometimes the comment stated the correct answer to the puzzle, but the script returned something else since it had a bug.
Neat! My mind-blown moment with GPT-4 was realizing that it will often be able to tell you the output of the (unique, not available in training data) scripts it writes for you.
It is impressive, but no, it won’t. It’ll tell you what a typical output for the command might be, and the more complex the script, the more wrong and full of hallucinations it will be. There’s a huge difference. Specifically, you have no way of knowing the difference between accurate outputs and inaccurate outputs, without running the command yourself, making it largely worthless. Without access to environment, it’s…
It doesn't even have to be complex. Ask it about a program in a language that's not very popular and the odds that it'll completely screw up its answers is high.