With Great gAIn Comes Greater Security Issues - When ML Frameworks' Scale for Growth Incorporates Security Risks to Users' Cloud Accounts

Back to the list of Speakers and Sessions
Watch the stream Level: {"label"=>"Level", "Beginner"=>"Beginner", "Medium"=>"Medium", "Nose bleed"=>"Nose bleed"}

There are various Machine Learning/BigData frameworks that have become quite popular in the past year due to the release of ChatGPT. This sudden popularity has caused that the scale for growth in parallel computing comes first and leaves aside the implementation of security mechanisms in some of the frameworks' components. In this talk I will go over the research process that I performed on one of these frameworks in an AWS install, showing how it started as two vulnerabilities in a web dashboard and quickly became privilege escalation in an AWS account.