AI BriefingAI Security Roles Command Up to $250/hr as Labs Stress-Test Models
·2026.10.31 01:01
Key point
Offensive security and vulnerability research roles for AI model evaluation reached a peak rate of $250/hr in a recent market scan.
Details
- AI training platforms are hiring security professionals for contract roles to stress-test models, with 14 of 29 matched hourly listings offering top rates of $100/hr or more.
- The highest observed rate was $250/hr for offensive security and vulnerability research, with a median top rate of $90/hr across the dataset.
- Specific high-paying roles include Security Architect at EUR 159/hr, GenAI security eval up to $150/hr, cryptanalysis at $135/hr, and penetration testing up to $125/hr.
- The work involves reviewing attacks, writing eval tasks, and executing jailbreaks, distinct from full-time SOC positions.
- Context includes a Binance security incident involving a compromised DevOps laptop and $6.3B in losses, which Elon Musk linked to future autonomous AI agents.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Security experts blame frontier AI labs for skipping basic containment in recent agent incidents