Wirestock has raised $23 million to build a marketplace that supplies multi-modal training data — images, videos, audio, and 3D assets — to AI companies. The platform connects creators who want to license their work with AI labs that need high-quality data to train their models. The round signals growing demand for ethically sourced training data as AI companies face mounting legal and reputational pressure over how they acquire the content that powers their systems.
What Wirestock Does
Wirestock operates a two-sided marketplace. On one side, photographers, videographers, musicians, and 3D artists upload their work. On the other, AI companies license that content for model training. Creators set their own terms. They get paid when their content is used. And they retain ownership of their work.
The platform currently has over 2 million creators and processes millions of assets. The content spans photography, video footage, audio recordings, music, and 3D models — the multi-modal data that next-generation AI models increasingly require.
Wirestock also offers tools that help creators optimize and distribute their content across multiple stock platforms simultaneously. The AI data licensing business builds on that existing distribution infrastructure — adding a new revenue stream for creators who were already using the platform.
Why AI Labs Need This
The AI training data landscape is under siege. Artists have sued AI companies for using their work without permission. The Oscar ban on AI-generated performances reflects Hollywood's pushback. The "This is Fine" creator is pursuing legal action against an AI startup that used his artwork. And the broader debate over AI and creative rights is intensifying across every medium.
AI companies need data to train models. But scraping the internet without permission creates legal liability and public backlash. Wirestock offers a clean alternative. Every piece of content is voluntarily submitted by its creator. Every license is explicit. And every transaction creates a paper trail that protects both sides.
The demand is growing as models become more multi-modal. Text-only models could train on publicly available web content. Models that process images, video, audio, and 3D data need curated, high-quality content that text scraping cannot provide. Wirestock is positioning itself as the supplier for that transition.
The Competition
Wirestock is not alone in pursuing ethically sourced AI training data. Getty Images launched a program licensing its photo library to AI companies. Shutterstock signed deals with OpenAI and Meta. And startups like Origin Lab are building marketplaces for video game data used in world model training.
Wirestock's differentiator is breadth. Most competitors focus on a single content type — photos or video or audio. Wirestock offers all of them in one platform. For AI labs training multi-modal models that need diverse content types, a single supplier simplifies procurement.
The pricing model also favors creators. Wirestock says it pays creators a higher share than traditional stock platforms. Whether that share is sufficient to compensate for the value AI companies extract from training data is a broader debate that extends well beyond Wirestock.
The Creator Economy Connection
Wirestock sits at the intersection of the creator economy and the AI economy. Creators produce content. AI companies consume it. The question is whether creators get fair compensation — or whether AI companies capture all the value while paying pennies for the raw material.
The same tension runs through the AI writing debate. AI image tools like ComfyUI give creators control over AI-generated output. But the models powering those tools were trained on data that creators often did not consent to share.
Wirestock's model attempts to resolve that tension by making consent and compensation explicit. Whether the economics scale — whether AI companies will pay fair prices for licensed data when free scraped data is available — remains the central question.
What It Means
Wirestock's $23 million raise is a bet that the market for ethically sourced AI training data will grow as legal pressure mounts and AI companies seek to clean up their supply chains. The same enterprise customers demanding responsible AI practices from their vendors will increasingly demand that the models they use are trained on properly licensed content.
For creators, Wirestock offers something the AI industry has largely failed to provide — a way to participate in the AI economy on their own terms. Whether enough creators and enough AI companies adopt the model to make it a meaningful market will determine whether ethically sourced training data becomes the standard or remains a niche.
