Levan Ostaevi

"Major AI Laws" and how the clean data structure makes it accessible

Week 8 of our curator program, and contributions keep getting stronger and stronger. Points are being allocated for the first time, and curators are showing what's possible with structured data. Plus, one of our amazing curators covered how AI is helping historians read carbonized scrolls that sat untouched for 275 years. Watch the full call here 👇

How is artificial intelligence unlocking the lost secrets of the ancient world? A new proof-of-concept maps out the "Digital archaeology" ecosystem, featuring breakthroughs like the Vesuvius challenge and 3D segmentation. @geoprotocol geobrowser.io/space/0c747ad3…

@perplexity_ai Why no-one take them to court. They either buy perfect products to kill or they downgrade and insult what they show???? Where is: Visualelectric??

Announcing Personal Computer. Personal Computer is an always on, local merge with Perplexity Computer that works for you 24/7. It's personal, secure, and works across your files, apps, and sessions through a continuously running Mac mini.

@geoprotocol Thanks for the highlight, @Geo! Tracking global AI laws is a heavy lift, but having a perfectly structured, decentralized database makes it accessible for everyone.

AI regulation is moving incredibly quickly. Governments across the world have been scrambling to put guardrails on AI. → Video interview laws. → Risk-based governance frameworks. → Mandatory labeling for AI-generated content. Each law reshapes what companies can and can't do with the technology. These laws exist in scattered government PDFs, legal blogs, and buried legislative archives. There was no single place where you could see them all, side by side, with full context on who introduced them, what jurisdiction they cover, and exactly what compliance looks like. Thanks to one of our amazing curators, @LOstaevi, that’s no longer the case: geobrowser.io/space/41e85161… A catalogue of every major AI law, including: → Date enacted and jurisdiction → Compliance obligations → Regulatory authority → The legislators who introduced them This is the kind of structured, verifiable foundation that legal teams, policy analysts, and AI builders actually need.

We made an animated short film with AI. It made people cry "ROOTS", a story about family, fatherhood, and the things worth holding on to. It was built entirely inside Freepik, by a team, from script to final cut Here's how we made it 🧵

We’re just getting started! 🔥 Huge thanks to @geoprotocol for the recognition. If you want to see how we’re structuring the world’s AI and Health data, check out the latest publications!

@geoprotocol Honored to be part of this! Mapping the AI policy landscape has been an intense but rewarding challenge. Looking forward to what we build next with the Geo team!

Publications are picking up pace across Geo! AI: → Active grant programs catalogued. → Major AI policy documents and regulatory frameworks mapped. Health: → All major organs structured, with primary functions and key diseases/procedures. Special shoutout to @LOstaevi, @rufat31947337, and @AthinkT for their amazing work!

@geoprotocol 🫡

Structured knowledge doesn’t happen by accident. That’s where curators come in. 120 bounties are live and assigned, 15 submissions are already in review, and 6 mentors are guiding curators through the process. Plus, the first two publications just dropped in the health space: Thanks to @Severussnaped for working on Top 200 Proteins. Details here → geobrowser.io/space/52c7ae14… Also, @LOstaevi for publishing Top 200 Exercises → curator.geobrowser.io/bounty/52c7ae1… This is what structured knowledge in motion looks like.

Since going live last week, contributions have started across Health, AI, and Crypto. Current focus areas: → Crypto security history → The open-source AI ecosystem → Vaccine efficacy data → Structured disease intelligence with sources If you have deep expertise in these areas, apply to join the Curator Program: geobrowser.io/curator-program

Week 4 of the Geo Curator Program. GRC-20 is now frozen. The base primitives are locked. New bounties are live. Mentorship is rolling out. First structured contributions are landing on the graph. Here's what happened this week ↓

We’re watching AI agents cross a threshold. From answering questions to taking actions. Hundreds of thousands of agents are now interacting with each other. But here’s the architectural problem: The information they operate on has no native context layer. No clear provenance. No “this came from here.” No “verified through this process.” No traceable chain of reasoning. That worked for humans. We infer credibility through intuition, pattern recognition, reputation. AI systems need structured, machine-readable provenance embedded in the information itself. This isn’t about one tool failing. It’s structural. The internet was not designed to carry provenance. It was designed for human consumption. As agents begin to collaborate, transact, and act on information, that missing layer becomes critical. Geo is building that layer: → Knowledge with provenance baked in → Claims that show their sources → Information that carries its own context Collaborative intelligence requires collaborative knowledge infrastructure. We’re building it for what’s already starting to happen.

Help me win the Ultimate Drum Kit from Drumeo! wn.nr/77rNycW

@Meta why is it so hard to contact a support team? Or at least get answers on an issue reported from different channels? Are there any humans working in meta?