Safety Research Scholar - Human Frontier Labs

<h2><strong>About the Program</strong></h2> <p><strong>The Human Frontier Collective (HFC)</strong> brings researchers and scientists together to tackle consequential problems in AI. In partnership with Scale Labs, we are building a small cohort of researchers to work on the empirical foundations of AI safety.</p> <p>That work includes finding where models fail, checking whether safety controls hold up under pressure in realistic settings, and studying how AI agents (mis-)behave over long-horizon tasks. Scholars take open research questions like these and turn them into work that improves how frontier AI systems are evaluated and safeguarded.</p> <p>Each scholar owns a substantial part of a safety research project. You'll work alongside a mentor from the research question through the experiments to deciding what to test next.</p> <h2><strong>Research Areas</strong></h2> <p>Mentors will guide Research Scholars on projects in selected AI safety research areas in select AI safety research areas, such as:</p> <ul> <li><strong>AI control and chain-of-thought monitoring:</strong> Developing techniques to oversee and constrain AIs, including detecting harmful intent or deception.</li> <li><strong>Alignment auditing and evaluations:</strong> Building scalable methods to test whether models have hidden goals, misaligned behaviors or unsafe tendencies.</li> <li><strong>Multi-agent safety: </strong>Studying the risks that emerge when multiple agents interact, such as collusion, coordinated failures, or unexpected collective behavior.</li> <li><strong>Recursive Self-Improvement (RSI) safety: </strong>Investigating how to keep AI systems safe and aligned as AIs increasingly contribute to their own development.</li> <li><strong>Science of Evaluations: </strong>examining when and how models recognize they're being tested, and how situational awareness affects the reliability/validity of evaluations.</li> <li><strong>Misuse risks and evaluations:</strong> Developing methods to asse