"Voice of Founders" From @playarts_ceo — Co-founder & CTO of @HoloStudio_AI “Decentralized infrastructure does not become real because a team says it is decentralized. It becomes real when people actively participate in building, testing, validating, and growing the network together. For HOOTI to succeed, we need real users willing to: 🔸 Run HOOTI nodes 🔹 Participate in QA 🔸 Test the protocol 🔹 Help bootstrap the network early The game ecosystem is simple: "PEOPLE PLAY". But decentralized AI infrastructure is different 👉 A sustainable protocol cannot depend on a single company forever. The network itself must become stronger than the founding team behind it. That is why onboarding matters now because distributed systems only become resilient when enough independent participants exist. If you believe open and community-owned AI infrastructure should exist, this is the stage where participation matters most.” — @playarts_ceo

The visual language of @HootiBrowser reflects this direction intentionally. As AI research becomes increasingly autonomous, the underlying infrastructure must become more open, verifiable, and collaborative. Below are the key premises that the Founding Team is aiming for ⤵︎

AI-generated research just crossed a major threshold. Sakana AI demonstrated that an AI-generated ML paper could pass peer review at an ICLR workshop, while the broader AI Scientist work was later published in Nature. This is not simply about replacing scientists. It signals that the research loop itself is becoming programmable. As AI agents begin generating hypotheses, running experiments, writing code, evaluating outputs, and drafting papers autonomously, the core bottleneck shifts toward infrastructure: 📍Who runs the compute? 📍Who verifies the experiments? 📍Who owns the research graph? A new research stack is emerging. Read more in threads ⤵️

3/ We have seen this pattern before. YouTube, TikTok, and Instagram did not create human talent from nothing. They changed the discovery layer. People outside TV networks, agencies, and traditional gatekeepers could suddenly reach the world from their rooms. The same shift is now coming to science and research. Humanity is not short on ideas. We are short on networks that can discover, test, verify, and compound them.

4/ That is the layer HOOTI + MAGNET is building. MAGNET turns research into parallel hypothesis loops. HOOTI is designed to route those loops across distributed nodes, record experiment provenance on-chain, and verify results through Byzantine-tolerant mechanisms. AI science is crossing the peer-review threshold. Now research needs infrastructure that is open, verifiable, composable, and not owned by a single cloud. 👉Paper: arXiv 2504.08066 👉MAGNET: hooti.ai

1/ The signals are converging. @sama keeps pointing at compute scarcity. @GoogleDeepMind is pushing distributed training forward with Decoupled DiLoCo. @karpathy’s AutoResearch showed that research itself can become an executable loop. The next AI stack needs a different shape.