Intelligence, the company behind the AI ranking platform Design Arena, announced a $7.9 million seed round on Monday. Index Ventures led the round, with participation from Conviction, A*, and Valkyrie. Sarah Guo and Mike Vernal represented Conviction in the deal.
The company started just a few weeks before graduation in 2025. Grace Li, co-founder of Intelligence, said the idea emerged from a college project. She and her friends were trying to build an AI game engine.
The models could produce functional games, but none were fun. The team found no substitute for human judgment when evaluating what made a game enjoyable. They brainstormed ways to get honest human feedback at scale.
"It was the missing bottleneck for a lot of these models to make improvements in the design space," Li said.
The timing proved fortunate. "About a week later, we closed our first major deal with a frontier lab, and the rest is kind of history."
How Design Arena Works
Design Arena functions like a model router with a ChatGPT-style window. Users see two outputs from media-generating models and choose between them. They rank the results from best to worst.
The platform has attracted 5.3 million users worldwide. On the enterprise side, it provides endless instant feedback for media-generating models. Users appear indifferent to which models they rank. They simply want the best output.
The site is currently generating $60 million in annual recurring revenue. That figure solidifies Design Arena's position as a key source of human-led evaluation data for the AI industry.
Users must log in to receive output. This requirement allows the company to track taste changes across continents and over time. Regional differences emerge clearly. Web dashboards in Asia, for instance, tend to feature a more maximalist design style than those favored elsewhere.
The Value of Human Judgment
The real value of Design Arena comes from its enterprise side. Media-generating models need feedback on design quality, and automated systems often miss the mark. User rankings give critical input into what people actually want.
Automated benchmarks remain scalable, but they are vulnerable to gaming and manipulation. Last week's breach at Hugging Face illustrated that risk clearly. The incident showed how easily automated evaluation systems can be compromised.
Stay ahead of the AI curve
The most important updates, news, and content — delivered weekly.
No spam. Unsubscribe anytime.
Human-led evaluation data has become a critical input for the AI industry. Design Arena's user base spans the globe, and its preference data reflects genuine human taste. That data helps frontier labs refine their models in ways automated tests cannot.
A Market With Mixed Results
The crowdsourced human feedback space has produced both failures and successes. Yupp, a startup in the same broad category, shut down earlier this year after raising $33 million from a16z crypto's Chris Dixon. The company had amassed over 1.3 million users but could not build a sustainable long-term business.
Yupp's collapse suggests that crowdsourced human feedback is not automatically a winning market. Building a durable company in this space requires more than just user volume.
Other startups based on human evaluation seem to be thriving. LM Arena, which takes a similar approach to text-based responses, raised $150 million in a Series A in January. That round came just four months after the company launched its paid product.
The contrast between Yupp's shutdown and LM Arena's success highlights the uneven terrain. Design Arena's $60 million ARR suggests its model has found firmer ground.
What Comes Next
Intelligence has not disclosed how it plans to use the new funding. The company's trajectory so far has been rapid. Forming the company, landing a frontier lab client, and reaching $60 million in ARR all happened within a short window.
The seed round brings in backing from established investors. Index Ventures led the deal, with Conviction, A*, and Valkyrie joining. The participation of Sarah Guo and Mike Vernal, both prominent AI investors, signals confidence in the approach.
The company's origin story remains tied to that initial insight about fun. The founders started with a simple observation: AI could make games, but not enjoyable ones. That observation led to a platform now used by millions.
Design Arena's growth reflects a broader shift in how the AI industry evaluates quality. As models improve, the bottleneck moves from generation to evaluation. Human judgment, once seen as a stopgap, now looks like a durable asset.
The company tracks preferences across continents, building a picture of global taste. That data has become valuable to labs trying to improve their media-generating models. The enterprise side of the business feeds this loop continuously.