LinkedIn is moving from advertising AI-training opportunities to becoming part of the infrastructure that supplies them. Its new AI Trainer Program uses the network’s identity, career and skills data to match professionals with projects that evaluate and improve artificial-intelligence systems.
A labour marketplace built on professional identity
LinkedIn says the program is being rolled out gradually while it gathers feedback. Any member can express interest, but the platform encourages applicants to verify their profile with government identification, discuss their expertise with an AI Project Matching Assistant and complete a written assessment of core trainer skills.
The model plays to LinkedIn’s central advantage: it already holds structured information about education, licences, employment and professional networks. For projects in law, medicine, finance or software, the company says those signals may be combined with an AI-powered conversation to evaluate fit and match people to relevant work.
“AI Trainers may review and evaluate AI outputs, provide expert feedback, share insights from real-world workflows, or contribute knowledge that helps improve AI systems.”
LinkedIn AI Trainer Program FAQ
What applicants are being assessed on
LinkedIn’s general assessment spans five categories: evaluation, preference ranking, rubric work, supervised fine-tuning and agentic work. A candidate might judge whether a model answer is accurate and safe, compare two outputs, write criteria for a good response or explain how an agent should handle a multi-step task.
The assessment is designed to test annotation fundamentals rather than deep subject knowledge. Domain expertise is evaluated separately through the member’s profile and professional conversation. LinkedIn says hiring teams can view structured results alongside education and experience, while integrity signals such as large copy-and-paste events may also be recorded.
Reported rates are attractive — but not universal
Business Insider reported that LinkedIn had published more than a dozen trainer listings during early testing. Some finance and nursing opportunities offered rates up to $100 an hour, while a senior software-engineering role listed a ceiling of $150 an hour. A red-teaming role was reported at $40 to $50 an hour.
Those figures describe specific public listings, not a standard pay scale. LinkedIn’s own help material says compensation varies by client and project. Work may be a one-time task or arrive in waves, and additional assignments depend on project demand, qualifications, availability and the quality of earlier submissions.
“Future work is not guaranteed, and some tasks may be assigned on a first-come, first-served basis.”
LinkedIn AI Trainer Program FAQ
A direct challenge to specialist platforms
LinkedIn enters a market shaped by companies such as Mercor, Surge AI and other expert-data providers that recruit people for frontier-model evaluation. Its advantage is distribution: hundreds of millions of professional profiles and an existing system for jobs, identity signals and recruiter search.
The program could reduce the cost of finding niche professionals while making AI-training work visible to people who would never join a specialist annotation platform. It may also make reputation more portable if completed projects lead to future opportunities or can be described, within confidentiality rules, on a member’s profile.
The contractor boundary matters
LinkedIn is explicit that trainers are independent contractors, not employees or representatives of the company. Applicants need valid work authorisation, may face background checks and are generally responsible for their own workspace and equipment. Completed tasks enter a review process before payment, which the company says typically reaches a bank account within 10 to 14 business days after acceptance.
That structure brings familiar questions from the wider data-labelling economy: how rejections are appealed, whether task time is fully compensated, what contributors may disclose and how sensitive professional knowledge is protected. LinkedIn’s trusted career identity may raise expectations for transparency and due process above those associated with anonymous gig marketplaces.
LinkedIn can make AI training feel like a mainstream professional category rather than hidden piecework. The opportunity is substantial, but the programme’s credibility will depend on whether verified expertise is matched with reliable projects, clear standards and compensation that reflects the value of the knowledge being transferred.