>>110131702
Bros is this a vibeGOD victory or not?

File: IMG_1722.png (1.46 MB, 1037x1058)

1.46 MB PNG

>>110131716
>that last line
Yes it is, we vibeGOD won

>>110131716
im literally sitting on 4 Windows 11 Criticals, 2 Apple High/Crit, and a Sony PS5 [[[VERY]]] Critical.
All unpublished.
I could not do this without Claude.
What a world.

>Accidentally

Some people just want to see the world burn, since they can't get a job or house. Shit world syndrome leads to even shittier conditions.

File: Screenshot 2026-10-22 at (...).png (49 KB, 898x503)

49 KB PNG

>>110131734
Yep. The PHP vuln poster was gloating.

>>110131702
> samba shares
shit nobody uses.
> local escalation
nobody else my moms basement except me
99% of these CVEs are non-issues.

>>110131702
Underflow, race, stack overflow, out of bounds.

C gods keep winning.

>>110131742
Cope more trans

>>110131702
Nobody gives a fuck.
None of those super bad vulnerabilities ever result in anything.

Meanwhile the one vulnerability, that got half the world attacked and caused the creation of a whole different type of malware (randomware), was literally a NSA backdoor that got leaked.
https://en.wikipedia.org/wiki/EternalBlue

File: images.jpg (18 KB, 509x392)

18 KB JPG

>>110131736

Anyone notice the closer we get to yet another AI company's IPO, the more doom advertising we see?

>>110131734
>Some people just want to see the world burn
well, they need to do a better job. just finding these isn't enough.

>>110131744
>Underflow, race, stack overflow, out of bounds.
um, have you considered, just, not writing those bro?

>>110131779
I mean its all real published CVEs man.
Things changed.

>>110131758
we say tranny here, you subhuman turdie retard

>>110131797
>published CVEs
Oh nooo, some stupid authority assigned a number to something.
The world is collapsing fr fr no cap!

Message me when something affects me.

>>110131742
wtf you talking about?
I use more SMB than NFS in 2026.
no one has time to set up NFS4 correctly, but SMB3 (even after you tard wrangle all the upcalling garbage for Kerberos + Dfs) is relatively easy.
KSMBD? well ya, no one wants that because even the authors know it's low quality C trash code full of holes.

>>110131831
anon things are good enough and as mentioned in another thread jail time v. bug bounty isn't worth it so shit goes straight to source devs for patching.
but you can't seriously deny published white papers documenting and providing deets on explicit replicatable zero days that pwn systems through and through.
i know its fucking gross and insane but things are what they are.

>>110131850
>you can't deny blablabla
Watch me do it.
None of this affects me.
Watch me not get hacked.

>>110131855
you wont but its because the system turns out is kind of regulating itself. happy for you, and me, for real.

>>110131798
You lost and AI CHADS won

>>110131850
>you can't seriously deny published white papers documenting and providing deets on explicit replicatable zero days that pwn systems through and through
Where are they?
We still don't know what the PHP thing yesterday was even about. No proof-of-concept. Just "trust me bro".
Get fucked.

>>110131728
You didn't do anything but lie.

>>110131861
oh really?
I got two dedicated servers right here, with php fpm and nginx running, and i will specifically not update them.

Surely one of those high skilled AI agents scanning the internet will find it and hack it, right?
If it doesn't get hacked, i consider the outrage of yesterday as fake and gay.
If you want to be taken serious, then do not lie!
If you cybersecurity tards keep lying, then don't complain that nobody takes your serious anymore.

>>110131876
Post public IP. thats it.
Will return results in less than 120 minutes.

>>110131886
Why would i have to post a public IP?
Those AI agents scrape me a thousand times per day. With a decent internet connection, you can scan the whole IPv4 range in no time.
If they don't get hacked, i consider it A LIE.
The CVS score was 10.0.
It doesn't go any higher. The claim is that anything running it on the public internet can get hacked.
If it doesn't get hacked, you LIED. It's not just blown out of proportions and "totally a real vulnerability, but just affects edge cases", but A LIE.

>>110131886
So you claim that you are one of those selected few unknown people, who knows what the unpublished PHP CVE, that has no proof-of-concept, was about?

Then why keep your secret?
Tell us what it is and share it with us.

>>110131909
>Why would i have to post a public IP?
lmao this guy is scared

>>110131779
the only vuln that meant anything in the past week was the jeetcert debacle (caused a couple of iot jailbreaks) and that was not ai (the reporter said so)

>>110131765
>was literally a NSA backdoor that got leaked
And it had a CVE score of 8.8, but got whole governments and corporations hacked.
Meanwhile yesterday we had a 10.0 and there isn't even one single known victim.
Nobody got hacked by it.

There are over 50 Million servers out there, still running the vulnerable PHP version.
Aren't we supposed to see something by now?

>>110131736
The fact that this supposed "lmao, i just want to see the world burn" dude didn't use the exploit to hack wikipedia or 4chan or do other cool things, but instead only did a video for twitter, tells you that it is a nothingburger.

>>110132010
it takes months for companies to report breaches

>bro just guess my IP haha
grim. Seriously the state of this board is grim. I mean at least entertain us, nignog definitely won’t be able to do shit.
You retards can’t prove a point to save your life.

File: 1792484848546214.jpg (9 KB, 230x220)

9 KB JPG

>It's SMB again

>>110131736
>Ive always been 'unethically ethical'.
Chaotic, then.

we already had php vulnerability and now this one again? are earth deadass?!

>>110131765
>randomware
it's so random UwU XD

>>110132010
>Aren't we supposed to see something by now?
all your personal data is already for sale in russia

>>110131702
why does it read like his biggest complaint is not the vulberabilities, but the fact that they weren't privately shared on the Chosen People's list before the goys could be warned?

>>110131716
If you're a developer working on your own, these tools at this point make your software better.

I ask Claude on Max effort to hunt for bugs, behavior that would subvert user expectation, duplicate functions, etc. it's quite good at finding things you would never have found ever, and writing endless fuzzers.

>>110132098
That's already the case since 2017. The Russians got access to all the CIA backdoors that time.
The biggest threat to national security is glowfags and their backdoors.
If you care about cybersecurity, shoot your local FBI / NSA / CIA agent in the face.

>>110131736
>fun
Understandable.

>>110132081
*pulls out spork*

>>110132041
Nigger, some 2b2t minecraft clans scan the whole IPv4 range multiple times per day for minecraft servers. And you can access shodan, if you are too lazy.
This is not the 90s anymore.

It's astonishing that cybersecurity idiots, who get hyped about the current fake and gay CVE, don't know the most fundamental basics.

You are all liars and scammers.

>>110132106
Did you use Fable or Opus?

>>110132162
*pulls out buttplug*

>>110132041
i self hosted a test minecraft server open to the internet, it took 30 minutes for a scanner bot to log on it
there are only 4.3 billion ipv4 addresses you fucking tardo

>>110131702
SMB. fucking kek.

>>110131798
you forgot to post frog picture /pol/ nignog

>>110131702
AI gods won again as usual

>>110131702
>accidentally
Yeah, suuuuuuure.

>>110132165
absolute schizophrenic delusional babel.
there is literally no way to point blank with even 0.000001% confidence determine an IP or pointer for an attack entirely blind with absolute no information whatsoever other than a verbal taunt.
what do you want me to do kike, scan all existent Minecraft servers apparently? All NA/EU CIDRs?

day of the rope. directly challenged, forwardly confident none of this is real, wont spin up a VM on some cloud service slop to see what happens. ok

File: luddite.jpg (173 KB, 1977x1092)

173 KB JPG

>>110131716
big W

and anon was full of shit as usual

>>110131716
No.
It's shit made by people that understand that LLMs make mistakes and engineered around it.
Ask for a million exploits, ask to create "proof of concept codes", run all the codes, and the ones that actually do something (1 in 1000) are re-checked and you have a nice big bucket of shit to fix or use.
Vibecoders don't understand how programming bugs work.
Of course if there was a second step on vibecoding where you use an LLM array like the one above to fix all the issues, the code would be great.

>>110132222
https://www.shodan.io/search?query=php

There are 741,126 servers within the USA alone, that you can hack right now in this very moment, with IPs known and listed there.
And you don't even have to scan for it yourself. I even entered the search term "php" for you, in case you are too lazy.

I expect you to run a botnet with hundreds of thousands of zombies within the next hour.
Because that is possible, according to the CVE tards and according to OpenAI.
You didn't lie, did you?

you morons aren't going to like what this psyop is about to do next

Winchads win again. You can't spell windows without "win".

>>110132555
only competent model will literally outright refuse on 247 adjacent inclusion strings to PoC.

>>110132576
>please don't hack me, here hack another php server

>>110132599
I imagine that a simple "please check the code for mistakes such as potential race conditions, vulnerabilities and code that will take too long to execute" post step would make the slop code a lot less sloppy

File: smug-cat4.png (2.11 MB, 1200x1199)

2.11 MB PNG

REMINDER that if there is no verifiable proof of concept, it is a lie.
Do NOT believe corporate or government employed security tards. They lie and deceive.

>>110132576
entirely non-answer, non-response, DNE - got it lol.
Too jeeted to even front a fucking AWS or DigitalOcean instance to even begin to pretend to stand on business.

>>110131716
This is a closed source victory

>>110132628
The slop bot would make up stuff and find issues that aren't issue.
And when you then prompt it to fix it, it will bloat your codebase with, at best, redundant garbage that does nothing.

You can't just tell it to "find race conditions", this will never work.
Think about Linus Tech Tips vibe coded patch set for the kernel, that didn't do synchronisation and introduced race conditions that humans spotted immediately.
The slop bot couldn't do that.

>>110132642
>>110132604
When the page cache thing got found in the kernel, there was a proof of concept repository and you could check it yourself.
When EternalBlue happened, we got flooded with reports of hacked businesses and government.

Meanwhile when the worst CVE of our lifetime happens, a RCE for everyone who runs php, with a score of 10.0 (now already downgraded to 9.2) and OpenAI promising a major severity.... Nothing happens?

>>110132677
We'll see, but this shit isn't even remotely as bad as log4j or react2shell were a few years ago. Log4j in particular was a complete suck fest.

>>110131734
you misunderstand. these vulnerabilities are intentionally inserted by nsa niggers. they get published once someone else discovers them so they get patched, to ensure that the us glowniggers are the only ones with backdoors. they have more.

>>110131765
>was literally a NSA backdoor
they all are

>>110132730
>w-w-we will see!
They will keep downgrading its score slowly.
Nobody is talking about a 10.0 anymore.
Then they will reveal that the exploit only works when you drink lemon tea during a full moon while running PHP on a Dell Optiplex from 1996, but only if you chose the Pentium 90/120 variants and turn it on by pressing its power button for exactly 1.527 ms, turning in a circle, clicking the heels of your ruby slippers together three times and repeats: "There's no place like home".

It is lies. It is always lies!
>b-but it's based on a real story!
Doesn't change that it's a lie.
And it is deliberate.
They don't do this accidentally. I don't believe that they are so fucking retarded that they AGAIN and AGAIN overhype total non-issues.
If you lie and lie, maybe at some point people will require that you prove your claims, before they believe you.

It's a lie until proven otherwise. Give me a proof of concept. Give me a repository.

>>110131765
the NSA is entirely responsible for 99.75% of CVEs btw

>>110132817
>Then they will reveal that the exploit only works when you drink lemon tea during a full moon ...
>but chud branch predictor what if...
>it wont

good, it's about time human written slopware got cleaned up
people were getting away with writing shitty code for decades because there wasn't enough manpower to pentest everything

>>110132785
>to ensure that the us glowniggers are the only ones with backdoors.
nb. this policy doesn't actually work. but they keep doing it.

>>110131702
so where the hell is it
post it my negra

File: 1769465711383789.png (223 KB, 512x468)

223 KB PNG

>>110131736
>Jλckλl

>>110131716
>Supreme intelligence
>Can't even report vulns correctly

>>110131797
Just like the vim CVE that has been known about for years and also a big nothingburger?

File: tympääntynyt tuijotus.jpg (449 KB, 1280x898)

449 KB JPG

>>110131702
> integer underflow
> rce

How is this even possible? We need to just delete every C program already.

>>110133027
>How is this even possible?
It's not, its a lie.

The current Rust sudo-rs privilege escelation however, is not.
https://ubuntu.com/security/notices/USN-8708-2

Closed source gods won
It's harder for AI to find vulnerabilities in a closed source program than an open source program. That means I can use AI on my source code to patch more vulnerabilities faster than some other random nigger using AI on my published binary can find them.
Open source is dead ahahahaha freetoddlers btfo

File: ghidra-logo.jpg (193 KB, 1155x635)

193 KB JPG

>>110133094
>closed so-

>>110131702
Here we go again...

>>110133101
first of all ghidra sucks and ida is better
secondly it's still an extra step and more time consuming to decompile and find bugs in a closed source program than to use the proper source code. I never said it was impossible, it's just a force multiplier in my favor compared with just releasing everything on shithub and letting any random faggot take pot shots without going through the effort to set up a proper decomplation work flow

File: natalia lurks.jpg (81 KB, 678x1000)

81 KB JPG

>>110133086
> Overview
> sudo-rs could be made to run programs as an administrator

>>110133162
You now remember that those are the people who complained about sudo being soo insecure and needing to be memory safe... just to produce yet another vulnerability every month.

If you have sudo-rs installed, you might as well just run everything as root, no difference.
Rust troons truly are the most incompetent retards this world has ever seen.

>>110131716
Yes. AI can bug hunt where its unprofitable or makes no sense for programmers to.

>>110131728
buy an ad

>>110131742
>> samba shares
>shit nobody uses.
wrong but like any microsoft "service" I would never deploy it on the public internet unprotected, if ever

File: 27a1ca8c-ac74-40ba-bd45-8(...).jpg (23 KB, 894x480)

23 KB JPG

>>110131797
it's just a card bugnutz

>>110133120
There's a massive torrent of kids doing reverse engineering for that "recomp" slop sea of shit.
LLMs make it trivial to reverse engineer shit at a point complete rertards can do it.

>>110133388
didn't the decomp platform shutdown because of all the non-working low quality slop nobody verified?

>>110131702
Good. Security through obscurity should be relegated to the dustbin of history.
Developers have a duty to the users, to disclose vulnerabilities to the public and not hide problems.

>>110133435
Yep!
But there's a big difference between getting a game to run in a different platform and just finding some random ass exploit.

>>110133194
Star craft brood war, and diablo 3 were the peak of computing.

>>110133301
The bigger danger is on LAN networks. Now any exposed Ethernet port = a known suite of exploits can fuck up any computer network running samba on an old version of windows (literally all of them.)

>>110134586
Its a linux kernel vuln not a windows one

>>110131728
why do pedo-cyberlarp-troons all have anime pfp?

>>110135763
cute and funny

>>110132631
why are you like this what made you so untrusting

>>110132577
Will it get me a gf?

>>110131779
yes, it’s been long-established that AI alarmism is their primary marketing strategy for pumping their market cap

it’s now also a strategy for trying to ram through legislation for making it illegal to run local AI or otherwise compete with Anthropic

File: 1792430657436532.jpg (156 KB, 750x1195)

156 KB JPG

>>110135763

>>110137652
why are you against making the internet more secure

>>110137652
This is stupid. Established by whom? Are you saying none of this security stuff is even real because AI found it? That's retarded, what's the point of such stupidity? Are you a seething chinaman?

>>110137529
you're hopeless. you might as well commit suicide, because you're soon to troon out

>>110138424
I'm too full of naturally occurring test for that. You wouldn't get it

>>110131702
I literally called this in the other thread where someone asked what was next and all you faggots said Poothon and Javascript.

>>110132104
Are you retarded?

>>110131702
The php vuln was 20 years old, 90%+ of vulns reported are fake hallucinations. So far, the "bombshell" vulns that were found "with AI" were all "if you have this driver enabled, which nobody has ever had enabled, not even 30 years ago, and then deliberately let input from the internet hit the driver, which nobody would ever do because this driver is unrelated to any of that stack". Total made-up nonsense.

>>110134575
Diablo 3 was total garbage slop 0/10 unplayable
>defending chink-tier gacha real money bullshit
>on my /g/

>>110138251
Have you actually tried looking at the AI reported vulnerabilities? Virtually all of them are completely made up and aren't even remotely vulnerabilities.

>>110138251
The php vulnerability definitelly isn't real.

>>110132959
Jthrembockthrembol

>>110137679
kek

>>110132051
>>It's SMB again
Olé, olé.... >>>/wsg/6257218

>>110131728
all you do is eat hot chip and lie

>>110132631
>if there is no verifiable proof of concept, it is a lie
you have the PoCs linked in the article, you absolute retard

>>110133086
>>How is this even possible?
>It's not, its a lie.
we got an expert here!

>>110135763
anime website

>>110131728
>4 Windows 11 Criticals
>look inside
>it's a "if you delete system32 the system crashes" situation
many such cases with clanker-made CVEs.

>>110139740
This. Had a cvss 5.3-scored "if the user calls this code, it could throw an exception, and if the user doesn't catch the exception, the user code will crash" CVE. Fix committed was to wrap it in try/catch and throw inside the catch block...

>>110139712
Are you literally lying about what is written in an article now?
This is getting ridiculous.
If you cybersecurity tards wouldn't lie so much, people would trust you more.

There is no proof of concept. There are claims that there would be one, and those claims refer to a twitter video where someone types:
>$ hack this
> Output: Uhh, i got hacked now

This is retarded. Give a proof of concept, like everyone else does, or fuck off.
You don't deserve a benefit of the doubt.

File: exploitvideo.png (72 KB, 768x452)

72 KB PNG

>>110131728
>What a world.
Don't forget the most important point to make your "unpublished" garbage look intimidating.

>>110139901
I found this: https://github.com/78ResearchLab/PHP2Shell-Full-Chain-Exploit
AI generated "exploit". Would that work? I don't care to test it. I do have a wordpress site and I updated because why not, but my impression is that it's yet more bullshit fake CVEs that are impossible to exploit in practice, if they're even real. Doesn't mean the use-after-free shouldn't be fixed though.

File: cat thinking.jpg (168 KB, 1776x1168)

168 KB JPG

>An open source project which is a not for profit idealist and mostly has no backing company or revenue claim has a submission process like a for profit company would in a project that needs security vulnerability to be kept secret or lead to financial loss.
I'm starting to think these wouldn't have been fixed if they weren't submitted this way

>>110139933
Why are you so profoundly retarded?

>>110131728
And Im literally sitting on your mom's face. She's licking my anus.

>>110139901
There used to be a saying, PoC or GTFO.

>>110139932
According to that vibe slop, you can only trigger that bug if you go out of memory while doing a multipart upload.
If a website would allow you to cause the server to run out of memory by an upload, that website would be unusable. So this exploit only works if you already have a different exploit.

nginx, apache and php itself all individually limit your max file upload size

A total nothingburger.

>>110139986
>According to that vibe slop, you can only trigger that bug if you go out of memory while doing a multipart upload.
Yes, it's what the reports I've read all say. Not only that but it only works if you make assumptions about memory layout and software stack, which are basically impossible in practice, due to cloudshit and load balancers. It also wouldn't work against php running in lambdas or something similar. If you limit file upload size it also doesn't work, and that's already default behavior. Etc. etc. etc. Hence why it's virtually impossible to exploit anyway. Even the research teams that looked into it said it only works in lab conditions with already-broken environment setups and wouldn't work on live internet sites.
But that there isn't even ONE (1) PoC for this is really cherry on top for me. Even with lab conditions only you should be able to generate a PoC if it's real.

>Planned fix in 7.5 if he hadn't leaked it
What is that even in, two years?

>>110131742
>> samba shares
>shit nobody uses.
I'd like to think everyone wishes this were true. At this point though the alternative is fucking sharepoint 365.

Holy shit man. Normies should have never been allowed to use computers.

>>110140543
he's partially right
nobody uses the kernel smb driver since its fucking SHIT
only cunts that use it are hardware companies that are spooked by GPLv3 samba

>>110140571
mount -t cifs doesn't call the mount syscall with the kernel smb driver? That's how I mount our "Y drive" at work.

>>110140920
It doesn't use ksmbd (the server and shitty part), it uses the cifs module.

description:    VFS to access SMB3 servers e.g. Samba, Macs, Azure and Windows (and also older servers complying with the SNIA CIFS Specification)
license:        GPL
author:         Steve French
alias:          smb3
alias:          fs-smb3
alias:          fs-cifs
import_ns:      SMBDIRECT
srcversion:     6A29C45A4B1F2A65B305195
depends:        smbdirect,netfs,dns_resolver,aead,libarc4,cifs_md4,nls_ucs2_utils,smb_compress

I have yet to see a CVE that actually does anything

>>110131742
>>110141258
>4chan's resident losers and sex offenders btfo by chads
>again
lmao