Got 99 Problems but Prompt Injection Ain't Pineapple
Explores the ethical challenges of AI vulnerability disclosure, covering data bias exploitation and responsible reporting standards.
Summary
Vulnerability Disclosure in Artificial Intelligence
Disclosing vulnerabilities in AI models presents unique hurdles compared to traditional software, requiring a balance between public transparency and preventing widespread exploitation.
Data Bias and Model Manipulation Vectors
Adversaries exploit algorithmic biases, data poisoning, and model manipulation techniques that can amplify vulnerability impact across automated decision-making systems.
Case Studies in Recent AI Disclosures
Evaluating real-world AI vulnerability disclosures highlights inconsistencies in vendor responses, patch validation, and public reporting mechanisms.
Refining Responsible AI Disclosure Frameworks
Establishing unified industry standards for AI vulnerability reporting fosters public trust and ensures consistent remediation across machine learning deployments.
Related talks
Creator Stage·30 min
Suha Hussain
Demonstrates incubated ML exploits using LangSec principles to inject backdoors into machine learning pipelines via model serialization bugs.
Main Stage·45 min
Bruce Schneier
Bruce Schneier examines how artificial intelligence systems hack human institutions, tax codes, financial markets, and legal frameworks.
Creator Stage·22 min
D. Jurado · J. Nogue
Chronicles the deployment of XBOW, an autonomous AI offensive agent that discovered over 200 open-source vulnerabilities and submitted 1,000 bug bounty reports.