Understanding the Threats and Attacks on Data Science Applications

Explore the critical security challenges facing data science applications and models in this 33-minute conference talk by Abraham Kang. Delve into the biggest blindspots in enterprise security and examine the extensive attack surface of data science applications. Learn about various attack vectors, including threats to algorithm data, infrastructure vulnerabilities, and risks to data science engineers and their development tools. Investigate potential weaknesses in data cleansing and preparation logic, as well as processing components. Address input problems unique to streaming applications and discover strategies for ensuring data boundaries. Gain insights on filter bypass issues, mismatched character sets, and the importance of understanding all potential data uses. Conclude with a Q&A session to further enhance your understanding of securing data science applications.

Understanding the Threats and Attacks on Data Science Applications
Biggest Blindspot in Enterprise Security
The Attack Surface of a Data Science Application
Attacking an Algorithm's Data
Securing Data Science Infrastructure
Attacking Data Science Engineers
Attacking Data Science Developer Tools
Attacking Data Cleansing and Preparation Logic
Processing Components
Input Problems Unique to Streaming Apps
Ensuring that Your Data is Bounded
Filter Bypass Issues
Mismatched Character Sets
Knowing All of Your Potential Data Uses
Questions and Contact