Latest AI Security Institute Jobs

Control Red Team - Research Engineer/Research Scientist

This role involves designing and running machine learning experiments to evaluate the effectiveness of AI control measures, such as monitors and sandboxes, by developing adversarial attacks and conducting security analyses. The work combines empirical research with real-world testing of frontier AI systems, using LLMs to automate evaluation loops and building scalable infrastructure for high-quality, reusable results. The position is embedded within a red team focused on identifying vulnerabilities in AI safety systems to inform both developers and government policy.

AI Security Institute London, United Kingdom

Software Engineer - Cyber & Autonomous Systems Team

This role involves working with cutting-edge technologies to evaluate the capabilities of AI models in cybersecurity and autonomy. Responsibilities include building evaluation infrastructure, developing analysis tooling, leading testing exercises, and delivering production-quality Python code. The work has a direct impact on shaping the UK government's understanding of AI cyber capabilities and informing safety standards.

AI Security Institute London, United Kingdom

Senior Research Scientist - AI Safety

Lead high-impact research in AI safety, focusing on red teaming and evaluations for frontier models in high-risk domains such as cybersecurity and CBRN. Develop novel mitigation strategies, shape research direction, and contribute to publications and stakeholder engagement. Work at the intersection of technical depth and real-world safety impact within a small, high-agency team.

Faculty AI London, United Kingdom

Hybrid Permanent Clearance Required

Researcher, Recursive Self-Improvement Safety

This role involves conducting research and developing strategies to mitigate risks associated with advanced AI capabilities, particularly focusing on recursive self-improvement. Responsibilities include designing and implementing risk assessment, control measures, and safety protocols, as well as collaborating with other teams to ensure robust AI safety practices.

OpenAI London, United Kingdom

On-site Permanent

Senior Software Security Developer – Core Platform Services

This role involves designing and implementing core security features for a high-performance simulation platform, with a strong focus on authentication, secure coding standards, and cross-team collaboration. The developer will act as a security SME, conducting threat modeling, code reviews, and mentoring engineers across the organization. Work spans identity protocols, platform guardrails, and incident response support within a fast-paced, AI-driven engineering environment.

PhysicsX London, United Kingdom

Senior AI Operations Manager

This role involves managing the operational performance and reliability of enterprise AI platforms in production, with a focus on incident management, monitoring, SLAs, and DevOps practices. The individual will ensure operational readiness, support continuous improvement, and coordinate with distributed teams while applying principles of AI governance and responsible AI. The position requires strong technical leadership in cloud-based AI environments, particularly with Microsoft Azure AI and Co-pilot.

eTech Partners London, United Kingdom £90,000 – £100,000 pa

Remote

Senior Technical Program Manager (InfoSec), London

Lead the delivery of complex security initiatives in a fast-moving AI-driven drug discovery environment, bridging security strategy and engineering execution. Translate CISO objectives into technical roadmaps, secure AI models and pipelines, and build automated dashboards for risk visibility. Partner closely with ML and engineering teams across the Alphabet/Google ecosystem to balance security, compliance, and research velocity.

Isomorphic Labs London, United Kingdom

On-site Permanent

Anthropic Fellows Program, AI Safety & Security

This role involves conducting empirical AI safety research over a 4-month fellowship, working on projects such as scalable oversight, adversarial robustness, and mechanistic interpretability. Fellows receive mentorship from Anthropic researchers, access to compute resources, and produce public outputs like research papers. The program supports remote participation in the UK, US, or Canada, with optional co-location in London or Berkeley.

Anthropic United Kingdom

Hybrid Internship

Security Engineering Manager, London or Lausanne

Lead and scale a security engineering team at the forefront of AI-driven drug discovery, designing and implementing a robust, AI-first security architecture for a digital biology platform. Translate complex scientific and business objectives into secure, automated systems while enabling high-velocity research in a regulated, interdisciplinary environment. Collaborate closely with ML researchers, medical teams, and Google/Alphabet stakeholders to secure cloud infrastructure, AI workflows, and sensitive life sciences data.

Isomorphic Labs United Kingdom

Agentic AI Engineer (Defence & Security)

This role involves designing and implementing cutting-edge AI solutions for defence and security clients, taking technical ownership of projects, mentoring junior colleagues, and contributing to new business opportunities. The work requires a strong background in Python and agentic AI, with a genuine interest in the defence sector.

Cambridge Consultants United Kingdom

Senior Manager (National Security)

Lead and deliver high-impact AI transformation projects for UK government and national security clients, bridging technical AI research with operational mission needs. Manage end-to-end project delivery, develop strategic proposals, and mentor cross-functional teams in a fast-evolving, high-stakes environment. Focus on responsible AI deployment, red teaming, and building trusted advisory relationships with senior stakeholders.

Faculty AI London, United Kingdom

On-site Permanent Clearance Required

Senior Data Scientist (Safety)

Lead high-impact AI safety projects by designing novel evaluation frameworks and robust software architectures for frontier AI labs. Mentor technical team members and ensure scientific rigour across the project lifecycle while contributing to thought leadership through research and open-source contributions. Work at the intersection of national security and cutting-edge AI development to advance responsible deployment.

Faculty London, United Kingdom

Hybrid Permanent

Senior Manager (Safety)

This role involves leading the delivery of AI safety evaluations and red teaming projects, forming strong relationships with clients and technical partners, and advising on AI safety strategy. The position requires a deep understanding of applied AI safety, commercial consulting, and the ability to bridge technical research with strategic business challenges.

Faculty London, United Kingdom

Hybrid Permanent

Anthropic Fellows Program, ML Systems & Reinforcement Learning

Engage in a 4-month full-time empirical research project on ML systems and performance, mentored by Anthropic researchers. Work includes building CPU simulators, accelerator backends, or synthetic data pipelines using open-source tools and large-scale systems. Projects aim for public outputs such as papers, with access to significant compute resources and a collaborative AI safety research environment.

Anthropic United Kingdom

Hybrid Internship

AI & Algorithms Engineer (Defence & Security)

This role involves leading the development of advanced AI and machine learning systems for defence and security applications, including signal processing, edge AI, and information processing with transformers. The engineer will deliver data-driven solutions, mentor junior staff, and contribute to client-facing technical proposals and demonstrators. Work includes translating research into deployable systems, with a strong emphasis on technical ownership and innovation in high-stakes domains.

Cambridge Consultants United Kingdom

Hybrid Permanent Clearance Required