Research Scientist – Human Influence in London

Benefits:

• Access to API credits for auto graders/agentic workflows
• Large compute budgets
• Funds for human user studies

Seniority: Mid

Education: PhD

Visa Sponsorship: Available

Security Clearance: Required

Posted: 5 Nov 2026

About The AI Security Institute:

The AI Security Institute is the world's largest and best-funded team dedicated to understanding advanced AI risks and translating that knowledge into action. We’re in the heart of the UK government with direct lines to No. 10 (the Prime Minister's office), and we work with frontier developers and governments globally. We’re here because governments are critical for advanced AI going well, and UK AISI is uniquely positioned to mobilise them. With our resources, unique agility and international influence, this is the best place to shape both AI development and government action.

The deadline for applying to this role is Sunday 29th November 2026, end of day, anywhere on Earth.

Team Description:

The Human Influence (HI) team focuses on the ways in which AI can influence or have a decisive negative impact on decisions made by humans or other AI systems. A substantial class of AI risk operates through social influence on people or other AI agents. AI systems can persuade others to change their beliefs and to take action; can build trusting relationships with people in order to exploit them; can extract private information from them; and can hold delegated ownership of high-stakes decisions. These can play out in both human-AI and multi-agent AI contexts.

Our work is highly interdisciplinary, drawing on methods from computational social science, AI safety and security, cognitive and behavioural science, machine learning, and data science. Typical projects include running rigorous human-AI interaction studies and randomised controlled trials, building evaluations and benchmarks that track AI capabilities across model releases, eliciting model capabilities through fine-tuning and self-play, and developing datasets to monitor real-world risk exposure and severity.

On Our Team, You Will Have:

• The opportunity to run large-scale studies, including RCTs, AI model evaluations, large scale data analytics, red teaming, and human participatory methods.
• The opportunity to engage with government officials, ministers, regulators, policymakers, frontier lab employees, and the wider research community to advance AI safety.
• Access to API credits for auto graders/agentic workflows and large compute budgets, as well as funds for human user studies.
• The opportunity to work with world-class talent.
• The opportunity to lead-author and co-author academic publications, conference papers, preprints, blog posts, and to disseminate work via talks and posters.

Some Examples Of Our Work Include:

• The largest-ever study on the persuasive capabilities of conversational AI.
• A large-scale study on how people use and follow personal advice from AI chatbots.
• A longitudinal study on how anthropomorphic AI facilitates human-AI relationship building.

Role Description:

We are looking for an ambitious and exceptional Research Scientist to join the Human Influence team. This could be in a sub-team lead or independent contributor role. As a Research Scientist, you will lead and contribute to research programmes that shape our understanding of how frontier AI systems influence people. You will design and execute rigorous studies, publish your work visibly, and help set the research agenda for the team. We are looking for someone who is technically strong, scientifically ambitious, and entrepreneurial, i.e., comfortable working at pace on high-stakes policy-relevant questions.

Who We're Looking For:

We are seeking candidates with experience in one (or ideally more) of these areas:

• Computer science, Machine learning, Statistics, AI research
• Computational social science
• Data science, especially including natural language processing
• Human-computer interaction
• Psychology
• Cognitive science

Essential Skills And Experience:

• PhD (or equivalent years of research experience) in a relevant field
• Track record of productivity, including publishing rigorous, high-impact research
• Experience designing, planning, conducting or analysing human experiments (ideally online and at scale)
• Strong knowledge of advanced statistical modelling methods (e.g. multilevel/hierarchical/mixed effects GLMs)
• Knowledge of frontier AI models and how they are trained and evaluated
• Strong coding skills (Python and/or R), including experience using AI coding tools (e.g. Claude Code, Codex, or similar)
• Experience working on AI model or system evaluations and other AI safety projects
• Strong verbal and written communication skills
• Demonstrable interest in the human impacts of AI

Desirable Skills And Experience:

• Published work related to societal impacts of AI, evaluation of AI systems, multi-agent systems, or relevant work in a related field
• Experience fine-tuning language models
• Experience with reinforcement learning
• Experience with large-scale analysis of multi-turn conversations or chat logs (e.g. using LLM-judges or auto graders)
• Front-end software engineering skills to build UI for studies with human participants

Nationality and Eligibility:

We may be able to offer roles to applicants from any nationality or background. As such we encourage you to apply even if you do not meet the standard nationality requirements.

Security:

Some of AISI’s work involves sensitive material and requires Security Check (SC) clearance. SC clearance is not required for this role; however, candidates who are willing and eligible to obtain it may have the opportunity to work across a wider range of projects. All candidates will be subject to the standard baseline security checks, including a Disclosure and Barring Service (DBS) check. Existing clearance status will not form part of the recruitment assessment.