Security Risk Analyst, Risk Engineering

Anthropic; New York (NY); Hybrid; USD 270,000 - 345,000; Full time

Job summary

Anthropic is seeking a Security Risk Analyst to help identify, quantify, and drive treatment of critical security risks across our AI systems. You will partner with Security Risk and Security Engineering to translate risk questions into decisions, using qualitative and quantitative methods with clear, defendable tradeoffs.

You will help shape risk appetite and standards, scale analyses with automation, and enable self-service insights for partner teams while preserving trust in our security

Qualifications

• Bachelor’s degree or equivalent in a relevant field.
• Experience with risk analysis in security or technology.
• Proven ability to quantify risk and present to leadership.

Responsibilities

• Enable leadership and partner teams to make risk-informed decisions with documented tradeoffs.
• Lead quantitative analysis of top security risk scenarios using FAIR and simulations.
• Assess threat scenarios with security engineering to align investments with actual risk.
• Frame risk treatments with clear recommendations and uncertainty, with triggers.

Skills

FAIR quantification

Monte Carlo

Python

R

Spreadsheet tools

Security risk analysis

Decision framing

Leadership communication

Education

Bachelor’s degree or equivalent

Tools

Python

R

Spreadsheet tools

Job description

About Anthropic

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.

About Anthropic

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.

About The Role

The Security Risk team is responsible for how Anthropic identifies, prioritizes, and drives treatment of its most important security risks. We are rebuilding risk management to operate as an engineering function, using automation, quantitative risk and AI-native platforms to enable decision making. The risks we assess span Anthropic's full security landscape, so the team needs breadth across domains and the judgment to go deep wherever a decision demands it. Security Risk, in deep partnership with Security Engineering, helps define the security program and shapes how investment and treatment decisions get made.

The conventional GRC playbook was not built for a company shipping frontier AI. You will help define what replaces it, with a direct line to CISO-level decisions. As a Security Risk Analyst you will take risk questions from leadership and partner teams and drive them to a decision. Sometimes that is a fast, structured qualitative assessment and other times it is a deep FAIR-based quantitative analysis. Either way you bring leadership a clear position, the tradeoffs, and honest uncertainty, and you defend it under challenge. In parallel, you will help turn quantitative risk into a product partner teams can leverage, and define the standards a growing risk function will run on.

Key Responsibilities

• Enable leadership and partner teams to make risk-informed decisions. Take ambiguous risk questions, drive them to a documented decision, and communicate the quantitative and qualitative tradeoffs clearly
• Lead quantitative analysis of the company's top security risk scenarios using FAIR, calibrated estimation, and simulation, working with the engineers who own the systems and presenting results in terms leadership can act on
• Partner with Security Engineering to assess threat scenarios and control effectiveness so security investment is right-sized to actual risk and remediation is sequenced where it buys down the most risk
• Frame escalations and risk treatment decisions with a clear recommendation, an honest statement of uncertainty, and re-evaluation triggers, and pressure test those decisions with risk owners
• Shape the analysis and narrative behind leadership risk reviews, and help define risk appetite and the risk metrics the organization should measure and why
• Use AI and automation to scale risk analysis, and own the calibration and quality review that keep the outputs trustworthy
• Turn one-off analyses into reusable methods, templates, and training so risk analysis becomes increasingly self service for partner teams, and raise the analytical quality of the risk register

You may be a good fit if you

• Have owned security or technology risk analysis end to end, qualitative and quantitative, and can point to a decision your output changed, whether a funding call, a launch call, a remediation sequence, or a documented acceptance
• Have hands-on FAIR-style quantification experience, including scenario decomposition, calibrated estimation, and Monte Carlo simulation in Python, R, or spreadsheet tooling, briefed to decision makers
• Thrive in ambiguity. You frame a moving question into something analyzable, pick the depth that fits, and drive to a decision rather than a report
• Put defensible severity and likelihood on ambiguous problems, state uncertainty honestly, and change your position when the evidence changes
• Have enough technical security depth to decompose an attack path with security engineers and be credible in the room
• Can compress a complex risk position into one paragraph for the CISO, make the tradeoff explicit, and defend it under pointed questions
• Use Claude or other LLMs as daily working tools and review model output critically
• Are low ego and collaborative, build credibility through the work, and care about AI safety and the role security risk plays in it

Strong candidates may also

• Have applied FAIR-CAM or another structured approach to control effectiveness and attack path modeling
• Have built or operated a cyber risk quantification program, or turned quantitative analysis into tooling, templates, or training that others ran
• Have helped define risk appetite or tolerance thresholds an organization actually used to make decisions
• Bring a background in security engineering, detection, threat intelligence, actuarial science, or decision science that grounds the numbers in real systems
• Are familiar with security risks specific to AI systems, such as goal or intent modification, and how they change traditional threat models

Annual Salary

$270,000—$345,000 USD

Logistics

Minimum education:

Bachelor’s degree or an equivalent combination of education, training, and/or experience

Required field of study:

A field relevant to the role as demonstrated through coursework, training, or professional experience

Minimum years of experience:

Years of experience required will correlate with the internal job level requirements for the position

Location-based hybrid policy:

Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.

Visa sponsorship:

We do sponsor visas! However, we aren't able to successfully sponsor visas for every role and every candidate. But if we make you an offer, we will make every reasonable effort to get you a visa, and we retain an immigration lawyer to help with this.

How We’re Different

We believe that the highest-impact AI research will be big science. At Anthropic we work as a single cohesive team on just a few large-scale research efforts. And we value impact — advancing our long‑term goals of steerable, trustworthy AI — rather than work on smaller and more specific puzzles. We view AI research as an empirical science, which has as much in common with physics and biology as with traditional efforts in computer science. We’re an extremely collaborative group, and we host frequent research discussions to ensure that we are pursuing the highest-impact work at any given time. As such, we greatly value communication skills.

The easiest way to understand our research directions is to read our recent research. This research continues many of the directions our team worked on prior to Anthropic, including: GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences.

Guidance on Candidates' AI Usage:

Learn about our policy for using AI in our application process.