Safety Research Scholar - Human Frontier Labs

Scale AI

4d • Remote

About The Position

The Human Frontier Collective, in partnership with Scale Labs, is building a small cohort of researchers to work on the empirical foundations of AI safety. Scholars investigate where models fail, whether safety controls hold up under pressure in realistic settings, and how AI agents behave over long-horizon tasks. Each scholar owns a substantial part of a safety research project and works alongside a mentor from research question through experiments and decisions about what to test next.

Projects are mentor-guided. Scholars are matched with a research proposal based on their interests and background, meet weekly with a Scale researcher for direction and feedback, and lead day-to-day work. Research areas may include AI control and chain-of-thought monitoring, alignment auditing and evaluations, multi-agent safety, recursive self-improvement safety, the science of evaluations, and misuse risks and evaluations. Projects are matched to demonstrated expertise; candidates are not expected to cover every area.

Scholars turn high-level safety questions into hypotheses, baselines, and controlled experiments; implement and debug evaluation harnesses, agent environments, datasets, monitoring methods, or post-training experiments; exercise independent research judgment; and produce reproducible code, experiment logs, analysis, and precise technical writing. Appropriate external outputs are subject to research, security, and communications review. Outputs include research papers intended for top machine-learning conferences such as NeurIPS, ICML, and ICLR.

The role is remote within the United States, including California and New York. It is structured as a part-time internship for 20 hours per week, initially for three months, with the option to extend to 6–12 months for exceptional candidates. Compensation is $100 per hour.

Requirements

• Motivation to ensure the safe and responsible development of advanced AI systems.
• Interest in technical safety research, such as alignment evaluations, AI control, monitoring, multi-agent safety, or open-weight model safety.
• Pursuing or having completed a PhD in computer science, machine learning, statistics, mathematics, or a related field, or an equivalent research track record.
• Ability to make progress with high-level guidance, including implementing, debugging, getting unstuck, interpreting results, and proposing next experiments.
• Clear research communication skills.
• Ability to handle confidential information and internal systems responsibly.

Responsibilities

• Own a safety research project from the initial question through hypotheses, baselines, controlled experiments, and decisions about what to test next.
• Implement and debug evaluation harnesses, agent environments, datasets, monitoring methods, or post-training experiments.
• Identify confounds, distinguish what a model can do from what it tends to do, interpret negative results, and determine next steps with high-level guidance.
• Produce reproducible code, experiment logs, analysis, and precise technical writing.
• Contribute research outputs, including papers for submission to top ML conferences, subject to applicable research, security, and communications review.
• Meet weekly with a Scale researcher for direction and feedback while driving day-to-day research work.

What This Job Offers

• Job Type: Part-time
• Career Level: Intern
• Education Level: Ph.D. or professional degree