Invisible acquires WeCP for expert credentialing

This deal pushes Invisible from supplying human labor into certifying which humans are qualified to judge AI. Invisible already breaks complex work into micro tasks for a 3,000 plus person global workforce, then sells that managed output to AI labs and enterprises. Adding WeCP brings a built in testing layer, so Invisible can screen, rank, and route specialists before their judgments feed model training, evaluation, or enterprise workflows.1 2 3

• Invisible’s core service is not just labeling data, it is workflow control. It routes model outputs to trained raters who score answers, rank alternatives, and annotate reasoning steps, while charging labs roughly $30 to $45 per hour and paying raters roughly $15 to $20 per hour. Assessment software strengthens that routing engine by making worker quality more measurable up front.1 2
• The strategic comparison is less traditional BPO and more AI labor marketplaces like Mercor. Mercor matches labs with pre vetted experts, while Invisible combines its own workflow software with a managed workforce. WeCP gives Invisible more of the screening and credentialing layer that marketplaces use to prove worker quality, but inside its own full stack delivery model.1 4
• This also fits Invisible’s move toward high precision, regulated work. Its research points to healthcare, finance, and defense as categories where buyers need auditability, traceability, and human accountability. In those settings, knowing that a rater passed a technical assessment matters as much as having enough raters on the platform.1 2 3

The next step is a tighter loop between testing, task routing, and model evaluation. If Invisible can use WeCP to continuously qualify experts and feed those scores into Meridial, it becomes harder to replace with a generic data labeling vendor, because the product shifts from labor supply to a system for producing trusted human judgment at scale.1 3