FutureJobPath logo
The career map for the AI era
This page explains how the Durability Score is built — the components, the evidence behind each one, and the named sources. For who this work fits and what a career path through it looks like, see the Deep Read. For your personalized match, take the free quiz.
Where the 51 comes from.

Three components - Automation Resistance, Structural Moat, and Demand - add up to 51.

Data note

Federal labor data does not count AI red-team engineers separately; the wage, workforce, openings, and AI-exposure numbers use Information Security Analysts as the public comparison. That fits the security base, but AI model-evaluation and release-review jobs are a narrower specialty.

FJP Durability Score
51/100
Automation Resistance
16/40

Automation resistance comes from threat judgment and severity calls, while routine test generation is easy to accelerate. AI can multiply prompt variants and organize failures, so resistance comes from threat design, severity judgment, and the ability to force fixes.

Sub-components
Substitution Resistance
6/30

Substitution resistance is limited for simple prompt testing, but stronger when the work designs realistic attacks and judges severity.

Sources feeding this sub-component
Anthropic labor-market impacts → Observed exposure for the Information Security Analysts occupation category is 48.59%.
Tufts American AI Jobs Risk Index → Median modeled job-loss pressure for the occupation category is 26.52%.
Augmentation Leverage
10/10

Augmentation leverage is high because AI can generate variants, organize failures, and help write repeatable tests.

Sources feeding this sub-component
IMF Staff Discussion Notes on AI and labor markets → Links AI-related skills with wage premiums in exposed labor markets.
Structural Moat
16/35

The moat is security credibility, reproducible findings, and engineering trust rather than a legal credential. The practical barrier is security credibility: reproducible reports, model-evaluation skill, engineering trust, and a history of helping teams remediate. The reports need practical fixes attached.

Sub-components
Physical & Environmental
1/10

Physical and environmental protection is minimal because the work is digital and remote-friendly.

Regulatory Moat
3/12

Regulatory pressure can help when buyers require documented model testing, but there is no protected license.

Robotics Resistance
8/8

Robotics are not the substitute path; the role is about security reasoning and model behavior.

Sources feeding this sub-component
Credential Depth
4/5

Credential depth is moderate through security experience, certifications, technical portfolios, and model-evaluation proof.

Sources feeding this sub-component
O*NET Online occupation summary → Lists this occupation in Job Zone 4, a higher-preparation category.
Demand
19/25

Demand has a strong security backdrop, but AI red-team roles are a specialty inside the broader information-security market. Hiring rides on the information-security market plus model-evaluation contracts, lab safety work, procurement scrutiny, and release review.

Sub-components
Volume
8/10

Volume is supported by strong information-security demand, though the AI red-team specialty is smaller than the parent row.

Sources feeding this sub-component
Bureau of Labor Statistics Employment Projections → Information Security Analysts: 182.8K jobs, 28.5% growth, and 16.0K annual openings.
Source Quality
6/8

Source quality is decent because information security analysts are a relevant public anchor, but not an exact specialty count.

Resilience
5/7

Resilience is fair because threat activity, procurement scrutiny, and model-release risk keep demand from depending only on hype.

What would move the score
Scenario 1
Prompt testing commoditizes

The case weakens if vendors automate large libraries of adversarial prompts and employers treat red-team work as a cheap pre-release checklist. That would compress entry roles that do not include security reasoning or remediation authority. That would turn shallow portfolios into weak evidence, because the market would expect deeper attack reasoning.

Direction
down
Components affected
Automation Resistance, Demand
Scenario 2
Evaluation becomes a buying requirement

The case strengthens if major customers, insurers, or regulators require independent model testing before deployment. That would create steadier demand for people who can document risks, reproduce failures, and verify fixes. Buyers would then need workers who can compare claims, verify fixes, and explain residual risk without hype.

Direction
up
Components affected
Demand
Scenario 3
Security and safety lanes split

A mixed outcome needs review if security-focused red teams grow while general safety prompt testing becomes temporary work. The important signal would be whether the role connects findings to product changes and accountability. The career risk is ending up in a narrow testing lane that never reaches security architecture or product authority.

Direction
neutral
Components affected
Automation Resistance, Demand
Personalized job matches →
Want to find the careers that fit your specific profile? Take the free FJP quiz — 3 personalized matches.
Last reviewed June 2026 · Next September 2026