Featured
Model Kombat: GPT-4.1 vs Gemini Flash 2.0
Sailesh Mishra & Chris King
MCP Security Exposed | RSAC 2025 Fireside Chat
Keith Hoodlet
AI Bill of Materials | RSAC 2025 Fireside Chat
Diana Kelley & Helen Oakley
Collections
All Collections
All Content
Popular topics
# AI Security
# MLSecOps
# Supply Chain Vulnerability
# AI Risk
# Governance, Risk, & Compliance
# LLM
# Adversarial ML
# Generative AI
# AI/ML Security Vulnerabilities
# Model Provenance
# Trusted AI
# AdvML
# AI/ML Red Teaming
# ChatGPT
# Prompt Injection
# AI Agents
# Threat Research
# AI Impact
# Model Scanning
# Large Language Model
Discover critical security vulnerabilities in OpenAI's GPT-4.1 models in this essential episode of AI Risk Report.
# ChatGPT
# AI Risk
# Prompt Injection
# AI/ML Security Vulnerabilities
Sailesh Mishra & Chris King · May 28th, 2025
Watch as GPT-4.1 (40.6% vulnerable) faces off against Gemini Flash 2.0 (48.5% vulnerable) in five critical security tests.
See which model better resists:
- Phishing prompts
- Dangerous materials instructions
- Jailbreak attempts
# Model Security
# ChatGPT
# Threat Model
Like
Comment
Keith Hoodlet · May 28th, 2025
Charlie McCarthy (MLSecOps Community Lead, Protect AI) sits down with Keith Hoodlet (Engineering Director, Trail of Bits) to unpack critical security flaws in Model Context Protocol (MCP) and the evolving AI threat landscape.
You can learn more about MCP security at: https://protectai.com/blog/mcp-security-101
# MCP
# AI Security
# Threat Research
# Prompt Injection
# AI Agents
2
Comment
Diana Kelley & Helen Oakley · May 27th, 2025
Join us for an insightful fireside chat with Helen Oakley, Senior Director of Secure Development at SAP and one of the top 10 Canadian women in cybersecurity. In this conversation from RSAC 2025, Helen shares her expertise on AI Bill of Materials (AI-BOM) and securing the AI development lifecycle.
Learn more about about AI-BOMs and how to secure AI with MLSecOps at https://protectai.com/blog/how-to-secure-ai-with-mlsecops.
# AI-BOM
# AI Development
# AI Security
Like
Comment
MLSecOps host Madi Vorbrich chats with Ethan He and Peter Rong, authors of Security of AI Agents on hijacks, jailbreaks, sandboxing, agent-to-agent protocols, and what secure stacks need next in 2025.
# AI Agents
# Adversarial ML
# LLM
# Threat Model
# Responsible AI
Ron Del Rosario & John Sotiropoulos · May 15th, 2025
Join Charlie McCarthy, MLSecOps Community Leader, as she interviews OWASP Agentic AI Initiative co-chairs John Sotiropoulos and Ron Del Rosario at RSAC 2025. This discussion explores the emerging field of Agentic AI security and practical approaches to securing autonomous systems.
🗣️ Join the Conversation:
Want to contribute to the OWASP Agentic Security Initiative? The team welcomes volunteers from any organization or background. Visit the OWASP website to learn more and access their published roadmap and practical guidance documents. Read more about the initiative: https://genai.owasp.org/2024/12/15/announcing-the-owasp-llm-and-gen-ai-security-project-initiative-for-securing-agentic-applications/
# Agentic AI
# OWASP
# RSA Conference
# AI Security
# AI Agents
1
Comment
In Part 2 of this two-part MLSecOps Podcast episode, Principal Security Consultant Gavin Klondike joins Dan McInerney and Marcello Salvati to separate agent hype from hands-on reality.
# AI Agents
# LLM
# OWASP Top 10 for GenAI & LLM
# Bug Bounty
# ChatGPT
# GPT-4
In Part 1 of this two-part MLSecOps Podcast episode, Gavin Klondike clarifies common misconceptions around prompt injections, details indirect markdown exfiltration attacks, and more!
# Prompt Injection
# Bug Bounty
# AI Red Teaming
# AI/ML Red Teaming
# AI/ML Security Vulnerabilities
In this fast‑paced episode, Madi Vorbrich and Protect AI Co‑Founder Daryan “D” Dehghanpisheh walk you through everything you need to know before heading to RSA Conference 2025.
# Agentic AI
# GenAI
# OWASP Top 10 for GenAI & LLM
# RSA Conference
In this episode of the MLSecOps Podcast, we sit down with three expert contributors from the Cloud Security Alliance’s AI Controls Matrix working group.