Mint Studio

The web is being rebuilt around AI agents coordinating with each other — and until now, nobody had actually measured how well that works. Together with Carnegie Mellon University, we've released AgentWebBench, the first benchmark designed to test decentralized agent-to-agent coordination across 100 websites and 18.4 million documents. We built this because we care about a healthy ecosystem, not just single-model demos, and the findings are more nuanced than the hype suggests. Decentralized coordination currently underperforms traditional search in most cases, but wins on factual Q&A and closes the gap fast as models scale. More importantly, the research surfaces real design principles for anyone building in this space: agents concentrate traffic on a small set of sources in ways that threaten open web diversity, planning matters more than raw model power, and we now have a proper framework for diagnosing where agent systems actually break down. This is important to @anaxilabs as we build our global data and agent supply chain for robotics and AI systems, when agentic AI is expected to become the coordination and abstraction layer for robotics. AgentWebBench turns abstract ecosystem concerns into measurable outcomes, and we think that's exactly what the industry needs right now. Full paper below. 👇

The economics of the agentic web — and why we built @mintstudio_ai. Full interview 🔎 technologyaiinsights.com/aitech-top-voi…

The window to own high-value AI assets is open right now — before agents start doing the selecting for everyone else. That's exactly what @mintstudio_ai is here for. Genesis Creator spots opening April 15. Follow and turn on notifications. 🔔

“Value accrues not just to whoever owns data, but to whoever gets selected by agents repeatedly.” —— Kate Shen, Co Founder of @anaxilabs This is the thesis Mint Studio is built on. If agents select the best prompts and workflows — repeatedly, at scale — then owning those assets is the most important move a creator can make right now. We're building the marketplace where that ownership is real, verifiable, and earns you revenue. Early access opening soon. 👇

The research says agents converge on the best sources. We're building the place where those sources live — and where their creators get paid. That's Mint Studio.

What a Powerhouse Collaboration! 🚀 @CarnegieMellon x @anaxilabs Taking the conversation from "How Platforms Survive" (as seen in our @Forbes report) to the new angle of 👉 "How Contributors Thrive" is a game-changer for the entire AI space. It's not just about how AI engines make money, but how we ensure data contributors get their fair share. 🙌

OpenAI is projected to lose $14 billion in 2026, with 800 million weekly users and 95% paying nothing. The model isn't broken. The monetization is. For her column in @Forbes , @forbes_sandra covers our new research in collaboration with Carnegie Mellon University that tackles the question the entire AI industry is avoiding: who actually pays for all of this? Our whitepaper, "An Economic Framework for Generative Engines: Advertising or Subscription," examines what happens when AI answers replace links — and what that means for the content creators, data contributors, and platform builders caught in the middle. The short version: extreme strategies fail in both directions. Platforms that go all-in on ads drive users away. Ad-free platforms can't sustain compute costs. The path forward is dynamic — and the economics need to be fair to work at scale. As our co-founder Kate Shen puts it: "Structural fairness is not just an ethical stance. It's a proven competitive advantage." Read Sandy's full breakdown in Forbes and dive into the full research paper. Both link in the comments👇 #GenerativeAI #AIMonetization #ArtificialIntelligence #AnaxiLabs #CMU

Proud to be partnering with our founders' alma mater, Carnegie Mellon University, one of the best schools in the world for computer science!