Testing Robustness Against Unforeseen Adversaries
OpenAI published a method to evaluate whether neural network classifiers can defend against adversarial attacks not encountered during training. The approach introduces a new metric called UAR (Unforeseen Attack Robustness) to quantify a model's resilience to unanticipated attacks. The work argues for measuring robustness across a broader, more diverse set of attack types rather than only those seen in training.
Related guides (3)
Related events (8)
Adversarial Attacks on Neural Network Policies
OpenAI published research examining adversarial attacks on neural network-based reinforcement learning policies. The work investigates how small, carefully crafted perturbations to observations can cause trained RL agents to fail catastrophically. This represents an early investigation into the robustness and safety of learned policies under adversarial conditions.
Computational limitations in robust classification and win-win results
OpenAI published research examining computational limitations in robust classification, exploring theoretical bounds on adversarially robust machine learning. The work investigates so-called 'win-win' results where both standard and robust accuracy can be achieved simultaneously. This is a foundational safety and robustness research contribution from 2019, addressing hardness results in adversarial ML.
Robust Adversarial Inputs: Multi-Scale Fooling of Neural Network Classifiers
OpenAI researchers created adversarial images that reliably fool neural network classifiers even when viewed from varied scales and perspectives. This directly challenges the assumption that self-driving car vision systems are robust to adversarial attacks due to their multi-angle image capture. The finding has implications for the security of deployed vision systems in safety-critical applications.
Transfer of Adversarial Robustness Between Perturbation Types
OpenAI published research examining whether adversarial robustness trained against one type of perturbation (e.g., L-infinity) transfers to other perturbation types (e.g., L2, L1). The work investigates the generalization properties of adversarial training across different threat models. This is an early safety and robustness research contribution from OpenAI predating the modern LLM era.
Strengthening cyber resilience as AI capabilities advance
OpenAI published a post outlining its approach to cybersecurity risk as its models grow more capable, covering risk assessment frameworks, misuse mitigation, and collaboration with the security community. The piece addresses both offensive risk (AI-enabled attacks) and defensive applications. It represents OpenAI's public positioning on responsible deployment in a high-stakes domain.
Attacking Machine Learning with Adversarial Examples
This 2017 OpenAI blog post introduces adversarial examples — inputs intentionally crafted to cause machine learning models to make mistakes, analogized to optical illusions for machines. It surveys how adversarial examples manifest across different input modalities and discusses the fundamental difficulties in defending against them. The post is an early foundational explainer on adversarial robustness from OpenAI.
How to Train Your Model Dynamically Using Adversarial Data
This Hugging Face blog post describes a methodology for dynamically training models using adversarial data, likely in the context of improving robustness against adversarial examples. The post covers techniques for generating and incorporating adversarial inputs during the training loop to improve model resilience. Published in mid-2022, it targets practitioners looking to harden ML models against distribution shift and adversarial attacks.
OpenAI Red Teaming Network
OpenAI is launching an open call for a Red Teaming Network, inviting domain experts to participate in ongoing safety evaluations of its models. The initiative aims to build a structured community of external red teamers who can help identify risks and failure modes across OpenAI's model releases. This represents a formalization of OpenAI's external adversarial testing program beyond one-off pre-release red teaming exercises.