SecÑinja⚔️

Security Engineer - (breaking and building app security) 👨‍💻 | Thoughts are my own.

[SecÑinja⚔️ @appexploit]

@kapsology wtf road minister so interested in fuel sceince? Can't he focus on fixing road first?

[SecÑinja⚔️ @appexploit]

@PrajwalTomar_ That’s why services like https://t.co/9m7O9pfVJk is there for who is serious about their business.

[SecÑinja⚔️ @appexploit]

@nullenc0de I would like to try this

[SecÑinja⚔️ @appexploit]

@JsonBasedman My request was sent to gpt2

[SecÑinja⚔️ @appexploit]

@ShivAroor Worst service. Driver never arrived

[Chaofan Shou @Fried_rice]

Chinese LLMs can hack better than state-sponsored hackers with properly evolved harness -

Kimi K2.5 managed to find and exploit 6 vulnerabilities in browsers: a single page view or an extension install by victims equal full system hijack.

Check https://t.co/d0SZSf1KqF

[SecÑinja⚔️ @appexploit]

@lukOlejnik Not really. Solving code security doesn’t mean it solves context and business usecases and logics as well as infra. Also, the human behaviour of leaking things here n there. On top that if AI secures everything then cybersecurity will majorly move to security of AI.

[SecÑinja⚔️ @appexploit]

@_jensec Interestingly I was thinking about solutions like Zeropath AI native code review platform. What will happen to them is the real question

[SecÑinja⚔️ @appexploit]

@_jensec I think on blackbox side XBOW will work like harnessing on top of the models like mythos. Wherein whitebox is straight forward just feed the code and you get the result. After a certain point whoever has efficient harness will Win.

[SecÑinja⚔️ @appexploit]

@gharkekalesh Controller disconnected

[Ihtesham Ali @ihteshamali]

This is wild 🤯

Someone built a tool that automatically removes safety alignment from ANY AI model.

It's called Heretic.

→ Works on Llama, Qwen, Gemma, Mistral and most major models
→ Runs on a consumer RTX 3090 in ~45 minutes
→ Achieves same refusal suppression as manually tuned abliterations
→ Preserves model intelligence better than existing methods

The scary part isn't the tool itself.

It's that every AI lab spending billions on safety alignment is now in a race against a 14-contributor open source repo.

Every safety update shipped → community figures out how to abliterate it → repeat.

This is what the AI safety debate actually looks like in 2026.

100% Opensource.

Link in the comments 👇

[AISecHub @AISecHub]

AI/ML Pentesting Roadmap - https://t.co/MfwmXU5ssZ

#AIMLSecurity #LLMSecurity #PromptInjection #AIPentesting #AdversarialML

[Ado @adocomplete]

Link to apply: https://t.co/dMJZ3Ed8YQ

[Om Patel @om_patel5]

THE MOST POWERFUL VIBE CODING PROMPT EVER:

[andrew gao @itsandrewgao]

you can instantly 10x your vibecoded frontends by just learning what different ui components are called

ofc opus is creating generic slop, the only words you know are menu and button.

[Phosphen @phosphenq]

this russian guy found a way to learn anything 10x faster🚨

saved 1,460h

NotebookLM + Gemini + Obsidian

dumps any source > AI strips duplicates > keeps only what u don't know yet

20 YouTube videos on the same topic
each one repeats the same 20% of information
this bundle removes the other 80%

what used to take a month > 15 min

yt: zproger

[Jerry @Mdhsan19]

Just dropped Jerry-list — my private bug bounty wordlist, now public and free for everyone.

Helped me land multiple P1s and P2s. DNS wordlist dropping soon too.

Go grab it and happy hunting.

https://t.co/TvCyduvlWN

#bugbounty #bugbountytips #infosec #recon

[Nicolas Krassas @Dinosn]

Google's Cybersecurity 2026 Forecast Report warns of a "Shadow Agent" crisis. These AI agents, deployed by employees without corporate oversight, can create invisible pipelines for sensitive information, leading to data leaks, compliance violations, and IP theft. https://t.co/0AvcspAgYY

[Nozz @NoahEpstein_]

stop what you're doing and look at this image.

each dot is 3.2 million people. 2,500 dots = 8.1 billion humans.

the grey? 6.8 billion people who have never used AI.
the green? 1.3 billion free chatbot users.
the yellow? 15-35 million who pay for it.
the red? that tiny sliver is us.

you think the AI space is crowded because you're in an echo chamber of the 0.06%.

the real world hasn't even started.

wrote a full breakdown on the data, the opportunity, and 7 businesses you can build from this gap today:

[mRr3b00t @UK_Daniel_Card]

A good collection of AI security resources:

https://t.co/xLNXfreSAS