OpenAI is reportedly asking third-party contractors to upload real examples of professional work they performed in current or previous jobs so the company can use those deliverables to evaluate and benchmark the performance of its next-generation AI agents, according to documents obtained by WIRED and first published on January 9, 2026.<\/p>\n\n\n\n
The initiative, facilitated in part through the training data company Handshake AI, requires contractors to provide two components for each task:<\/p>\n\n\n\n
OpenAI\u2019s instructions emphasize that submissions must reflect \u201creal, on-the-job work\u201d that the contractor has \u201cactually done,\u201d with examples drawn from long-term or complex assignments (hours to days of effort). Contractors are directed to remove or anonymize any confidential, proprietary, personally identifiable, or material nonpublic information before uploading, and the company provides a ChatGPT-based \u201cSuperstar Scrubbing\u201d tool to assist with redaction.<\/p>\n\n\n\n
Also Read<\/strong>: OpenAI Looks for New Leader to Spot and Stop AI Dangers- Head of Preparedness<\/a>
The goal is to create realistic, high-quality human baselines against which future AI agents can be measured for progress toward automating economically valuable office tasks, a key milestone on OpenAI\u2019s path to AGI. The documents describe hiring contractors \u201cacross occupations\u201d specifically to collect these real-world tasks modeled after their full-time roles.<\/p>\n\n\n\n