Published Aug 22, 2024

Threat modeling LLM apps

Dive into the world of LLM security with expert insights from Donato Capitella as he unpacks cybersecurity practices, tackles the challenges of threat modeling, and underscores the importance of output validation to safeguard applications against potential threats and vulnerabilities.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Collaboration & evaluation for LLM apps
Answers 383 questions
Capabilities of LLMs 🤯
Answers 383 questions
AI is changing the cybersecurity threat landscape
Answers 383 questions
Large Action Models (LAMs) & Rabbits 🐇
Answers 383 questions
The new AI app stack
Answers 383 questions
Only as good as the data
Answers 383 questions
Generative models: exploration to deployment
Answers 383 questions
Automating code optimization with LLMs
Answers 383 questions
Protecting us with the Database of Evil
Answers 383 questions
Deep-dive into DeepSeek
Answers 383 questions
Creating tested, reliable AI applications
Answers 383 questions
Testing ML systems
Answers 383 questions
Mamba & Jamba
Answers 383 questions
Machine learning at small organizations
Answers 383 questions
Controlled and compliant AI applications
Answers 383 questions

Dexa/Practical AI

Threat modeling LLM apps

Topics covered

Popular Clips

Securing LLM Use Cases

AI Alignment Challenges

Safeguarding LLMs

LLM Security Insights

Monitoring AI Security

Validating LLM Outputs

Alignment Challenges

Output Validation Strategies

Jailbreaking AI Models

Securing LLM Deployments

Model Efficiency Insights

LLM Security Insights

Future of AI Security

LLM Security Risks

Episode Highlights

Cybersecurity Practices

LLM Security Challenges

Output Validation

Related Episodes

Collaboration & evaluation for LLM apps

Capabilities of LLMs 🤯

AI is changing the cybersecurity threat landscape

Large Action Models (LAMs) & Rabbits 🐇

The new AI app stack

Only as good as the data

Generative models: exploration to deployment

Automating code optimization with LLMs

Protecting us with the Database of Evil

Deep-dive into DeepSeek

Creating tested, reliable AI applications

Testing ML systems

Mamba & Jamba

Machine learning at small organizations

Controlled and compliant AI applications

Threat modeling LLM apps

Topics covered

Popular Clips

Episode Highlights

Cybersecurity Practices

LLM Security ChallengesDonato Capitella explores the intricacies of threat modeling for LLM applications, emphasizing the need for robust security measures. He discusses the use of security canvases and strategies to mitigate prompt injection risks, ensuring safer deployments.

LLM Security Challenges

Output ValidationDonato Capitella discusses the importance of validating outputs from large language models (LLMs) to ensure secure operations. He explores various techniques and the potential impact of unsafe outputs on applications and users.

Output Validation

Related Episodes