Robert Kirk

Publications

Open Technical Problems in Open-Weight AI Model Risk Management

Robustness

Open-weight frontier AI models are becoming more capable and widespread, offering unique advantages and risks compared to proprietary systems. This paper identifies 16 technical challenges related to data, training, evaluation, deployment, and monitoring, and argues that true progress requires openness not just in model weights, but also in research methods and evaluations to build a rigorous science of open-model safety.

September 30, 2025
Date Range

STACK: Adversarial Attacks on LLM Safeguard Pipelines

Robustness

We tested the effectiveness of "defense-in-depth" AI safety strategies, where multiple layers of filters are used to prevent AI models from generating harmful content. Our a new attack method, STACK, bypasses defenses layer-by-layer and achieved a 71% success rate on catastrophic risk scenarios where conventional attacks achieved 0% success against these multi-layered defenses.

July 1, 2025
Date Range

News

Layered AI Defenses Have Holes: Vulnerabilities and Key Recommendations

Red-Teaming & Evaluation

We tested the effectiveness of "defense-in-depth" AI safety strategies, where multiple layers of filters are used to prevent AI models from generating harmful content. Our a new attack method, STACK, bypasses defenses layer-by-layer and achieved a 71% success rate on catastrophic risk scenarios where conventional attacks achieved 0% success.

July 1, 2025
Date Range

Research

Our research explores a portfolio of high-potential agendas.

Events

Our events bring together global leaders in AI.

Programs

Our programs build the field of trustworthy and secure AI