| ★ |  How To Hack AI (Exploiting insecure output handling in LLMs) | | 36 | — | |
| 7 |  Learn AI Fundamentals in 3+ Hours - Full Course | The Cyber Mentor @tcmsecurityacademy | — | Estados Unidos | 90 |
| 8 |  Why Large Language Models (LLMs) Output Differs | The Cyber Mentor @tcmsecurityacademy | — | Estados Unidos | 90 |
| 17 |  Google Big Sleep: The End of Human Hackers? | | 932.3K | Reino Unido | 88 |
| 5 |  OWASP's Top 10 Ways to Attack LLMs: AI Vulnerabilities Exposed | IBM Technology @ibmtechnology | 244.2K | Estados Unidos | 90 |
| 10 |  5 Claude Hacks So Powerful They Should Be Illegal | | 32.0K | — | 89 |
| 20 |  Free LLM that beats DeepSeek… can also generate videos!? | Futurepedia @futurepedia_io | 12.7K | Estados Unidos | 88 |
| 18 |  Prompt Injection in AI | Data Engineering @dataengineeringtamil | 12.6K | India | 88 |
| 6 |  LLMjacking: How hackers steal your AI API keys and stick you with the bill | IBM Technology @ibmtechnology | 11.5K | Estados Unidos | 90 |
| 15 |  LiteLLM Compromised | Abhishek.Veeramalla @abhishekveeramalla | 10.6K | India | 89 |
| 9 |  legally hack AI in 13 minutes | | 10.0K | — | 90 |
| 11 |  HACKERS Are Using This AI Tool | | 9.2K | — | 89 |
| 12 |  You Asked About AI: Agents, Hacking & LLMs | IBM Technology @ibmtechnology | 6.0K | Estados Unidos | 89 |
| 13 |  How to Pentest LLMs Like a Security Researcher Cybersecurity | | 4.2K | India | 89 |
| 19 |  Stop Overengineering: Workflows vs AI Agents Explained | What's AI by Louis-François Bouchard @whatsai | 1.7K | Canadá | 88 |
| 14 |  Hacking AI without code | What's AI by Louis-François Bouchard @whatsai | 1.0K | Canadá | 89 |
| 16 |  AI Security Skills Will Replace You if You Don't Learn Them Now! | CTRL+ALT+DEFEND @mpcybersecurity | 651 | Reino Unido | 88 |
| 4 |  Hacking and Securing APIs (OWASP API TOP 10) 2024 | | 120 | — | 90 |
| 1 |  How To Hack AI (Exploiting LLM APIs with excessive agency) | | 37 | — | 99 |
| 2 |  How To Hack AI (Exploiting vulnerabilities in LLM APIs) | | 25 | — | 98 |
| 3 |  How To Hack AI (Indirect prompt injection) | | 15 | — | 93 |