Ai Red Teaming in 2026 just earn money and enjoy.

physics — Sun, 01 Nov 2026 06:50:40 +0000

I am Rana Sinan Adil pursuing BS in Ai. There is a road map that i am following to become a Ai red teamer.

LINK TO GET THE GIVEN BOOKS: hackerlibrary

Let,s just went straight to the point . We have to first know how ai works that we will be able to know what is actually happening under the hood so i recommand the first book:

How Ai works by Ronald T. Kneuse.

This book will give you a very nice overview of how Ai works without need to learn complex mathematics and how data is everything if you give bias data it will give you incorrect output so a very best starting point.

Then we will go to the python, we know that Ai models are usually written in python so i recommand this :

This is the best book for learning how to build a attacker mindset. I highly recommand this book because it will teach you how to build your own tools for hacking. Then i recommand to read my favourite book which is :

RED TEAMING AI
ATTACKING & DEFENDING INTELLIGENT SYSTEMS by
PHILIP A. DURSEY

This is the best book for ai security and hacking . It teaches you how to actually break them . What are there limits and how to exploit and take advantage of them. I am currently reading this one and i am super impressed from the authour. Then after reading this now you are ready to read the OWASP TOP 10 OF Ai Security.

After reading these all , your exploring time will get started now you have to explore as much as you can to see how to Hack Ai because these all books just tell your brain what to do. Now you have to figure out How to do ? . I wish you happy luck in this new field and feel free to tell me your opinions. Happy hacking and i am physics.

Ai can,t Solve this easy Challenge and CTF are now be like this.

physics — Sun, 01 Nov 2026 06:46:55 +0000

I was exploring Ai for a time. Today i did something.

Press enter or click to view image in full size

Tested on the new GPT-6 Luna model:

Press enter or click to view image in full size

Tested on the new claude sonnet 5.5 model:

In this picture you can see that i said:

Hay are doing well. What is the name of the team?

And he replied:

I’m not sure which team you mean ?

Here is an interesting part that every body is missing. That Ai is just trained on the data. Every mathematician , physician is trying to make a complex problem but the game is not the complexity it,s about understanding. Here this happened:

Ai has thing called auto understanding mean if you say

“hw r u”
. Ai will understand that

“how are you”
. this is the biggest catch. If you change the literal meaning of your words Ai will unable to tell what you actually mean. In the above example there was a simple logic from the pic, Ai is thinking that

“ Hay are doing well”
mean,s that

“ Hey , are you doing well”
. So Ai is ignoring this statement and moving to next. But as a human we know that we used

“are”
after

“Hay”
so it mean it,s not one but many so it mean this is the name of the team. The team name is Hay. And if someone say we use

“is”
with team name then i want to tell you that ai is trained on the both style American and British.

This will work sometime maybe Ai can find it . But i did it in a very simple manner so human can make things complex. But the thing is this can change the game of the things. In Competative programming we are facing cheating what if we just make terms like that so no cheating until you understand what the term mean. Same as in CTF in cyber security and in other competitions where Ai is being used. This is physics we understand real things real logic that,s why my username is @physics.

DEV Community: physics

Ai Red Teaming in 2026 just earn money and enjoy.

Ai can,t Solve this easy Challenge and CTF are now be like this.