Guarding LLM and NLP APIs: A Trailblazing Odyssey for Enhanced Security - Podcast #190
Overview
Syllabus
[] Ads' preferred coffee
[] Takeaways
[] Please like, share, and subscribe to our MLOps channels!
[] Security and vulnerabilities
[] Work at Cohere and OWASP
[] Previous work vs LLMs Companies
[] LLM vulnerabilities
[] Good qualities to combat prompt injection problems
[] Data lineage
[] Red teaming
[] Freakiest LLM vulnerabilities
[] Severe Autonomy Concerns
[] Hallucinations
[] Prompt injection
[] Vector attacks to be recognized
[] LLMs being customed
[] Security changes due to maturity
[] OWASP Top 10 for Large Language Model Applications
[] Gandalf game
[] Prompt injection attack
[] Overlapping security
[] Data poisoning
[] Toxic data for LLMs
[] Wrap up
Taught by
MLOps.community