Trusted AI Blog

• LLM Security

How to perform LLM Red Teaming for ChatBots and test for JAilbreaks, Prompt Injections, and Data leakages like a boss.

This week in AI security: California's generative AI procurement guidelines, third-party testing as AI policy, and Treasury on AI risk in financial services.

This week in AI security: the Databricks AI Security Framework, a primer on hacking large language models, and a UN resolution on safe and secure AI.

• LLM Security

This edition covers the LLM security threat model and a serious incident, prompt injection techniques, LLM security courses, and resources for CISOs.

• GenAI Security

Despite efforts to encrypt communications, a newly developed attack has demonstrated the ability to decode AI assistant responses with alarming accuracy.

This week in AI security: over 100 malicious AI and ML models found on Hugging Face, conditional prompt injection in Copilot, and self-spreading AI worms.

In the ever-evolving landscape of AI within cybersecurity, 2024 brings forth profound insights from Mr. Alex Polyakov, CEO and co-founder of Adversa AI.

This scrutiny emphasizes the necessity for a steadfast commitment to Quality and Robustness testing before releasing AI in production.

• GenAI Security

This week in AI security: DARPA and IBM on protecting AI systems from hackers, a new book on generative AI security, and the OWASP LLM governance checklist.

• LLM Security

This week in AI security: Menlo Ventures on the modern AI stack, drawing the parallel with how quickly cloud platforms scaled for enterprise computing.

Data poisoning in machine learning datasets poses a significant threat, allowing attackers to manipulate model behavior intentionally.

• LLM Security

Welcome to the latest edition of our LLM Security Digest!