According to co-founder Grace Li, the idea behind DesignArena originated in 2025: a few weeks before graduation she and several college friends were trying to get their AI game engine to work. The models could produce functional games, but the outputs were rarely enjoyable — prompting the question of how to predict whether a game would be fun.
The team concluded there was no substitute for human judgment and began exploring ways to gather honest human feedback at scale. That work produced DesignArena, an interface now used by 5.3 million people worldwide.
How DesignArena works
For non-enterprise users the product resembles a sophisticated model router: a Chat-GPT-style prompt window with dropdowns for websites, images and a dozen other visual formats. After specifying a request, format and style, users are shown a series of “A vs. B” comparisons and asked to rank outputs from best to worst.
That workflow is useful on its own, but Intelligence’s core value proposition is on the enterprise side. Media-generating frontier labs can use the platform as a source of continuous, instant human feedback. Because users generally don’t care which underlying model produced an output — as Li puts it, they just want the best output — their rankings provide direct signals about real user preferences.
Funding and revenue
On Monday Intelligence, the company behind DesignArena, announced a $7.9 million seed round led by Index Ventures, with participation from Conviction (Sarah Guo and Mike Vernal), A*, Valkyrie and others.
Li says the site is currently generating $60 million in annual recurring revenue (ARR), which the company cites as evidence of its role as an important human-led evaluation source for the AI industry.
Why human feedback matters
Intelligence requires users to log in to view outputs, allowing the company to track how tastes vary across continents and over time. Li notes, for example, that web dashboards in Asia tend toward a more maximalist design style. Such human-centered measures complement automated benchmarks, which can scale more broadly but are often susceptible to gaming or manipulation — a risk highlighted by the recent Hugging Face security breach.
Market risks and comparisons
Crowdsourced human feedback is not a guaranteed business winner. The article recalls Yupp, which shut down less than a year after launching despite raising $33 million from a16z crypto’s Chris Dixon; Yupp claimed over 1.3 million users and had some frontier models as customers but failed to build a sustainable long-term business.
Still, other startups focused on human evaluation appear to be thriving: LM Arena, which applies a similar approach to text-based responses, raised $150 million in a Series A in January, just four months after formally launching its paid product.
Conclusion
DesignArena’s growth to 5.3 million users, Intelligence’s $7.9 million seed round and the firm’s reported $60 million ARR indicate strong demand for scalable human feedback that helps tune AI-generated media. At the same time, prior shutdowns remind the industry that translating user-labeled data into a durable business remains challenging.
