Safety Research Scholar - Human Frontier Labs

Job Overview

The Human Frontier Collective (HFC) is seeking a Safety Research Scholar to join their team. The HFC, in partnership with Scale Labs, aims to build a small cohort of researchers to work on the empirical foundations of AI safety.

Research Areas

The scholar will work on projects in selected AI safety research areas, guided by mentors. These areas include:

• AI control and chain-of-thought monitoring
• Alignment auditing and evaluations
• Multi-agent safety
• Recursive Self-Improvement (RSI) safety
• Science of Evaluations
• Misuse risks and evaluations

Responsibilities

As a Safety Research Scholar, you will:

• Own a substantial part of a safety research project
• Work alongside a mentor from the research question through the experiments to deciding what to test next
• Turn a high-level safety question into hypotheses, baselines, and controlled experiments
• Implement and debug evaluation harnesses, agent environments, datasets, monitoring methods, or post-training experiments
• Exercise research judgment to identify confounds, separate what a model can do from what it tends to do, and decide what to try next
• Produce reproducible code, experiment logs, analysis, and precise technical writing

Requirements

You may be a good fit if you:

• are motivated by reducing catastrophic risks from advanced AI systems
• are interested in technical safety research
• are pursuing or have completed a PhD in CS, ML, statistics, mathematics, or a related field
• can make progress with high-level guidance
• communicate research clearly and handle confidential information and internal systems responsibly