For most of the past decade, stock photography was a slow, predictable business. Photographers uploaded images, customers licensed them, platforms took a cut. Then AI arrived and the entire logic of that market broke.
Wirestock saw the shift early. It had spent years helping photographers distribute their work across platforms like Shutterstock. In 2023, it changed direction entirely and started selling that content to AI companies instead. This week it announced $23 million in fresh funding and reported $40 million in annual run-rate revenue. That is a real business, built fast, in a market that barely existed three years ago.
The data training market Wirestock is entering is currently worth around $3.5 billion globally and is on track to grow past $20 billion within the next decade. Image and video content already accounts for the single largest share of that market. Wirestock's specific niche, rights-cleared creative assets for AI models that generate images, videos, and 3D content, is sitting at the fastest-growing edge of it.
The legal pressure on AI companies is part of what is making this business possible. Getty Images sued Stability AI in 2023, alleging that the company scraped over 12 million copyrighted images without any license. That case is still working through courts in both the US and the UK. Disney and Universal recently filed against Midjourney over similar claims. Record labels have filed against AI music generators. The message from content owners is consistent: take our work without paying, and we will sue.
This legal environment is a direct tailwind for companies like Wirestock. AI labs that want to avoid litigation need data providers who can demonstrate consent, clear rights, and documented provenance. Wirestock's model, where artists knowingly opt in and get paid for their contributions, is precisely the paper trail that an AI company's legal team needs before signing a large contract.
The competitive context matters here too. The broader data labeling industry is already producing companies worth tens of billions. Scale AI was valued above $29 billion after Meta's 2025 investment. Surge AI generated over $1 billion in revenue in 2024 while operating without any external funding. These companies mostly focus on text, code, and human feedback tasks. Wirestock is positioning in a different lane: visual and creative content, where the rights picture is messier and specialist curation is harder to replicate.
There is a risk worth naming. Synthetic data, meaning AI-generated training content, is growing fast as a potential substitute for human-produced assets. Some analysts predict synthetic data will eventually outpace real-world data in AI training volume. But the current consensus among researchers and practitioners is that synthetic data works best as a supplement, not a replacement. Models trained exclusively on synthetic content tend to degrade over time as they lose contact with the messiness of the real world. For creative content specifically, where the whole point is human originality, the argument for real human-made assets is even stronger.
The more interesting long-term question is who else is sitting on a library of original content and has not yet thought about its value to AI companies. Retailers with years of product photography. Hospitality brands with architectural and interior imagery. Publishers with decades of illustration archives. Design agencies with proprietary visual assets. None of these organisations built their content libraries with AI licensing in mind. Some of them may be unaware that what they have is now in demand, rights-cleared and documented, by the same companies currently signing eight and nine-figure data contracts.
Wirestock's story is not really about a startup raising money. It is about a structural change in who holds value in the creative world and what that value is actually worth.