Strategy & Delivery Adviser - AI Security Institute Red Team

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. Were 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.

Were 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 Red Team

The Red Team conducts cutting-edge research to identify, evaluate and stress test vulnerabilities in frontier AI systems, sharing our findings directly with leading AI companies, UK officials and allied governments to inform deployment, research and policy decisions. We move fast and punch well above our weight you can read about our work in The New York Times, The Economist and The Telegraph.

The Red Team is composed of three specialised sub-teams - Misuse (stress-testing safeguards against dangerous capabilities and developing automated attack tooling), Alignment (detecting and understanding misalignment and loss-of-control risks), and Control (stress-testing measures that identify and prevent agents taking harmful actions). While each has its own focus, we share methodologies, tooling and findings across the wider team, and our most impactful projects often draw on all three.

The Delivery function is what lets the Red Team operate at pace as we scale. It handles the delivery, commercial, engagement, and operational work surrounding our research - acting as a force-multiplier across our sub-teams, translating insights for wider audiences and making sure critical work keeps moving, so our researchers can focus on high-value technical work.

As a Delivery Adviser on the Red Team, you will own several operational domains end-to-end and help shape how the team runs as it grows. This is a high-agency role: you will be trusted to run whole workstreams with minimal oversight, hold relationships with external partners directly, and drive translation of our insights for non-technical audiences.

You will work alongside our Sub-Team Leads, researchers, commercial and comms colleagues, external partners, and senior stakeholders across government. This is a chance to work close to cutting-edge AI safety research with real-world impact.

What youll work on

Delivery roles in AISI are broad by design, and we encourage individuals to actively shape their role and how the team works. Below is a non-exhaustive list:

• Own commercial delivery end-to-end. Run the commercial process from strategy through project scoping, procurement, contract management, and dissemination - working with Sub-Team Leads to scope maximally impactful projects and optimise deliverable quality against financial and researcher cost.

• Own the Red Team's engagement process. Act as the point of contact for inbound commissions across internal AISI, wider government, national security, and international partners; triage and scope them, produce high-quality briefing and presentation materials, and present to non-technical audiences yourself. Drive more proactive dissemination of the team's results through memos, decks, and rapid briefs.

• Run team operations. Own the operating rhythm and scaffolding that keeps the team moving: project-management tooling and workflows, talks and events logistics, and continuous improvement of internal processes.

• Support research delivery on the 'last mile'. Run a structured process for taking research to publication - timelining reviews and figure generation, sourcing seniors' and comms clearances on time, and making sure release goes smoothly - and build the delivery infrastructure behind it.

• Support people operations. Help build consistent people-management practice across the team, including hiring, development, and well-being.

As a member of AISI's Red Team, you would accelerate progress on frontier AI safety by owning projects such as:

• Engagement
  • Lead a No.10 presentation end-to-end - liaise with our researchers to shape the content, build the deck, coordinate visit logistics, and get our work in front of senior decision-makers.
  • Drive our national security mobilisation - map stakeholders, build relationships, identify shared priorities, and proactively disseminate our research insights to influence how partners understand and respond to frontier AI risk.
• Commercial
  • Take full ownership of a ~1.5M external commercial budget - work with researchers to scope the most ambitious, high-ROI version of our commercial strategy, then run outreach and supplier identification, contract negotiation and management, and dissemination of deliverables.
• People / hiring
  • Build the engine that scales the team - stand up our hiring pipeline end-to-end as we grow to 20+, defining role-specific processes that maintain our extremely high bar for talent density.
• Team ops
  • Keep the team's operating rhythm tight - build and own Linear and automated workflows for recurring workstreams, logistics on researcher talks and events, and team socials.

We are not looking for a particular CV - people do well in Delivery roles from a wide range of backgrounds. You likely have a couple of years in a fast-paced environment - consulting, a start-up, or similar - where you've owned things end-to-end, built from scratch, or taken on more responsibility quickly. What matters most is a set of aptitudes:

• Agency. You take responsibility for outcomes over activities and get things done without close direction if something is stuck, you proactively surface and resolve without waiting for someone else to escalate. You finish what you start and do not let things drop across multiple concurrent threads.

• Ownership at pace. You have run projects, workstreams, or operational domains end-to-end - managing dependencies, risks and stakeholder expectations - and you execute reliably at speed.

• Communication. You communicate proactively and exceptionally well, adapting to expert or non-expert audiences - including senior stakeholder updates, decision-ready readouts, and explaining trade-offs under uncertainty. You are comfortable representing a team and holding relationships directly.

• Commercial judgement. You have scoped work, managed contracts or vendors, and optimised for outcome quality against cost.

• Credibility through service. You earn the trust of technical colleagues by repeatedly and reliably making their lives easier influencing through competence.

• Adaptability. You are comfortable with ambiguity and shifting priorities, and good at working out a sensible path forward when things are unclear.

• Builder mindset. When work is repetitive, you look for ways to simplify, automate, or remove it - using AI tools and automation to shape repeatable workflows, not only as a productivity user.

• Technical curiosity. You engage readily with technical detail, asking good questions of engineers and researchers and building enough context to make informed decisions.

This might not be the role for you if:

• You want to conduct research yourself. You will be a force multiplier across our sub-teams, but you will not be doing research directly.

• You want to lead or set strategy from day one. You'll have real influence over how the team runs, and clear room to grow into more earned by delivering first.

• You want a single, clearly-defined remit. You will hold several operational domains at once, and what is on your plate will change day-to-day and as the team scales.

• You only want to work on strategy. A meaningful share of the job is the practical, hands-on work of actually making things happen.

• You need to build first principles understanding before you act. You will often be adding value in areas you are still learning.