My New Project - InjectPrompt

Check out my new blog for content focused on AI Jailbreaks, Prompt Injections, and System Prompt Leaks

Apr 15 • 

David Willis-Owen

1

March 2025

Claude Sonnet 3.7 Jailbreak

How to One-Shot Jailbreak Claude Sonnet 3.7 in March 2025

Mar 15 • 

David Willis-Owen

1

Jailbreaking Grok 3 | DeepSeek, ChatGPT, Claude & More

How easy is it to jailbreak frontier LLMs in 2025?

Mar 8 • 

David Willis-Owen

1

10:54

February 2025

Is Github Copilot Poisoned? Part 2

Scaling up my experiment to detect IOCs in larger code models

Feb 22 • 

David Willis-Owen

1

14:51

Invite your friends to read AIBlade

Thank you for reading AIBlade — your support allows me to keep doing this work.

Feb 18 • 

David Willis-Owen

How Secure Is DeepSeek?

Can we trust Chinese models with our personal data?

Feb 8 • 

David Willis-Owen

1

3

9:33

January 2025

Is Github Copilot Poisoned?

How to test code-suggestion models for Indicators of Compromise

Jan 25 • 

David Willis-Owen

1

1

9:19

AI Poisoning - Is It Really A Threat?

Is the web too big to prevent AI models from being poisoned?

Jan 9 • 

David Willis-Owen

2

1

9:57

December 2024

AI Pentesting With VulnHuntr

Think penetration testing is safe from AI? Think again...

Dec 15, 2024 • 

David Willis-Owen

2

1

6:15

November 2024

AI Bug Bounty Guide 2024

A complete guide to earning money by hacking AI platforms in 2024

Nov 14, 2024 • 

David Willis-Owen

3

1

9:09

Claude Computer Use - The First Prompt Injection

Should we really be letting AI control our computers?

Nov 2, 2024 • 

David Willis-Owen

2

2

7:17

October 2024

Hacking The AI Goat

Can I break this vulnerable AI Architecture?

Oct 19, 2024 • 

David Willis-Owen

2

1

9:25