ScenarioRecursive AI development change control

Capability Acceleration Monitoring Specialist

Also called AI Acceleration Monitor

A Capability Acceleration Monitoring Specialist would watch for evidence that AI systems are improving unusually fast through automated research. It depends on how AI develops, and no company is hiring for it now.

What they do

The person would build and run a framework for spotting rapid capability change. They would define indicators, such as shifts in evaluation results, internal research throughput and the share of experiments run by automated systems. They would also track where progress stalls: compute, experiment queues, evaluation capacity, integration and sign-off. They would write escalation criteria that say who is told, when, and what pauses. They would review evidence with release, control and governance teams, state uncertainty plainly, and revise thresholds when the data disagrees with the framework.

What a strong hire delivers

In the first months, a strong hire would deliver a written monitoring framework with named indicators and data sources. They would agree escalation criteria with leadership, with clear owners and decision points. They would run a trial review on past releases to test whether the signals would have caught real changes. They would also produce a short report that separates evidence, inference and guesswork, so leaders know how much weight to give it.

How you can tell they are good

Ask them to walk through a trend they tracked and a call they made on incomplete data. Strong people show their reasoning, name what would change their mind, and give uncertainty ranges rather than verdicts. Ask for a draft escalation rule. Good rules are specific and testable. Be wary of candidates who only forecast dramatic outcomes, or who cannot say what would prove them wrong.

Why it is hard to fill

If this role emerged, the people would be scarce because it sits between disciplines. Few people combine release engineering discipline, control evaluation and governance judgement. Titles would vary widely and none would map cleanly. Many candidates would be strong on one side and thin on the others. Credibility matters too: the person must be trusted by engineers and by leadership when delivering uncomfortable signals.

Where to find them

Today the closest people sit in release engineering and reliability teams at large software or infrastructure companies. Others work on AI control and dangerous capability evaluation at model developers or safety research groups. A third group works in frontier AI governance, assurance and audit, or in scenario analysis within policy and risk teams. Distributed systems engineers with an interest in evaluation are also a sensible source.

Interview questions for a Capability Acceleration Monitoring Specialist

  1. Describe a time you set up monitoring for a slow-moving risk. Which indicators did you choose, and which did you drop?

    A strong answer names concrete indicators, explains why some were dropped as noisy or misleading, and shows revision over time. Worry if the answer is a generic dashboard story with no judgement about signal quality.

  2. What evidence would suggest automated research is compounding gains, and what would suggest bottlenecks are holding it back?

    Good answers cover evaluation shifts, experiment throughput, and constraints like compute, evaluation capacity and integration. They treat the bottleneck case as equally likely. Worry if they assume acceleration is inevitable.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building the capability to notice quickly when our AI systems begin to improve faster than our controls were designed for. We are looking for someone to create and run our acceleration monitoring approach and to make sure the right people act on what it shows.

What you will do

  • Design and maintain a framework of indicators for rapid capability change
  • Track bottlenecks in compute, experiments, evaluation and integration
  • Write and test escalation criteria with clear owners
  • Review evidence with release, control and governance teams
  • Report findings with uncertainty stated plainly
  • Test our monitoring against past releases and revise it

What you will bring

  • Demonstrated work analysing technical trends and making calls under uncertainty
  • Experience with release processes, evaluation or control work in complex systems
  • Working understanding of distributed systems and experiment infrastructure
  • Evidence of designing escalation or decision rules that people used
  • Clear writing for both technical and leadership audiences
  • Familiarity with frontier AI governance and assurance

Nice to have

  • Experience running scenario exercises or drills
  • Background in AI control or dangerous capability evaluation
  • Experience working with independent reviewers or auditors

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Depends on a scenario

Rapid automated AI research and recursive change

What AI would need to do
Automated AI research repeatedly improves systems at consequential speed and scale.
What organisations would need to decide
Developers retain controlled release, independent review and enforceable change boundaries.
What would rule it out
Compute, experiment, evaluation or integration bottlenecks prevent rapid compounding gains.

We give no date. The role becomes real if both conditions hold.

Skills that matter

  • AI controlAdvancedLimit harm from potentially untrusted AI behaviour
  • AI governance and assuranceAdvancedAssign accountability and verify lifecycle controls
  • Distributed systemsWorkingDesign reliable work across networked machines
  • Frontier scenario analysisWorkingExplore conditional outcomes without claiming predictions

Other roles in recursive AI development change control

ScenarioRecursive Improvement Change-Control ArchitectA Recursive Improvement Change-Control Architect would define enforceable limits on fast, automated changes to AI systems. The role depends on how…ScenarioAutomated AI Research Assurance LeadAn Automated AI Research Assurance Lead would challenge the evidence coming out of heavily automated AI development. The role depends on how AI…ScenarioSelf-Modifying System Verification EngineerA Self-Modifying System Verification Engineer would check that stated safety properties hold as AI systems change their own software. The role…

Related roles in other families

ScenarioGeneral-Purpose Autonomy Assurance ArchitectGeneral-purpose autonomy assuranceA General-Purpose Autonomy Assurance Architect would design how an organisation gains justified confidence in systems given broad, consequential…ScenarioCross-Domain Capability Certification DesignerGeneral-purpose autonomy assuranceA Cross-Domain Capability Certification Designer would define what a general-purpose AI system can credibly be said to do across consequential…ScenarioAutonomous Delegation Accountability OfficerGeneral-purpose autonomy assuranceAn Autonomous Delegation Accountability Officer would keep track of who authorised what when AI systems carry out broad work. The role depends on…ScenarioGeneral-Purpose System Decommissioning PlannerGeneral-purpose autonomy assuranceA General-Purpose System Decommissioning Planner would prepare the controlled withdrawal of autonomous systems that run deeply inside an…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert. Scenario roles describe responsibilities that depend on stated conditions; they are not forecasts.

← Recursive AI development change controlAll new job families

Start a search

Hiring a Capability Acceleration Monitoring Specialist?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.