AIPwn

[paper] Prompt Injection 2.0 — The Hybrid AI Threat

What it is — and why it matters now

Sep 2 • 

aipwn

August 2025

[paper] Hacking the Hive Mind: How Multi-Agent LLMs Get Jailbroken

New research shows optimized prompt attacks can outsmart defenses like Llama-Guard

Aug 26 • 

aipwn

April 2025

I embarked on my AI Bounty journey

On April 1, 2025

Apr 1 • 

aipwn

November 2024

Black Friday Special: AIPwn Newsletter - Your Gateway to AI Security

Best Subscription Opportunity of the Year!

Nov 28, 2024 • 

aipwn

June 2024

[paper] MARKLLM: An Open-Source Toolkit for LLM Watermarking

we introduce MarkLLM, an open-source toolkit for LLM watermarking

Jun 2, 2024 • 

aipwn

April 2024

[paper]LLM4Decompile: Decompiling Binary Code with Large Language Models

Large language models (LLMs) show promise for programming tasks, motivating their application to decompilation

Apr 16, 2024 • 

aipwn

March 2024

[paper]Logits of API-Protected LLMs Leak Proprietary Information

Potential Information Leakage in API-Protected LLMs

Mar 18, 2024 • 

aipwn

[paper] ImgTrojan: Jailbreaking Vision-Language Models with ONE Image

"ImgTrojan: Jailbreaking Vision-Language Models with ONE Image," the introduction of a novel attack mechanism against Vision-Language Models (VLMs) is…

Mar 14, 2024 • 

aipwn

OpenAI Introduces Multi-Factor Authentication for AI Conversations

Is your OpenAI account safer now?

Mar 12, 2024 • 

aipwn

1

[paper] Here Comes The AI Worm: Unleashing Zero-click Worms that Target GenAI-Powered Applications

ComPromptMized: Unleashing Zero-click Worms that Target GenAI-Powered Applications

Mar 11, 2024 • 

aipwn

A Safe Harbor for Independent AI Evaluation

We make AI safer

Mar 5, 2024 • 

aipwn

[paper]Watermark Stealing in Large Language Models

In this paper, identifying watermark stealing (WS) as a fundamental vulnerability of these schemes.

Mar 5, 2024 • 

aipwn