OpenAI Reportedly Asks Contractors to Upload Real Work Files for AI Evaluation

Jan 11, 2026
OpenAI is reportedly asking contractors to upload examples of real work from past or current jobs to evaluate AI models’ performance. Contractors are instructed to remove confidential data before submission, according to Wired.

OpenAI is asking third-party contractors to upload examples of actual work they have completed in past or current jobs to help evaluate its AI models, according to Wired. The effort is reportedly part of OpenAI’s broader initiative to assess how well its systems perform compared to human professionals.

The company, together with Handshake AI, is requesting contractors to submit real work outputs such as Word documents, PDFs, PowerPoint presentations, Excel files, images, or code repositories. These submissions are meant to form a benchmark of human performance across various occupations.

OpenAI instructs participants to strip proprietary and personally identifiable information from their submissions. A tool described as “ChatGPT Superstar Scrubbing” is provided to assist with removing sensitive content. Contractors are also permitted to create realistic but fabricated examples if they prefer not to share original material.

Legal experts cited by Wired cautioned that this approach carries potential risks related to trade secrets and confidentiality. They noted that relying on contractors to judge what information can safely be shared could expose AI firms to intellectual property concerns.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like AI Policy Brief or Daily AI Brief.

Also, consider following us on social media:

AI Safety & Regulation AI Brief AI Brief (X)
Free newsletter

AI Policy Brief

Weekly report on AI regulations, safety standards, government policies, and compliance requirements worldwide.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.