rewrite this content using a minimum of 1000 words and keep HTML tags
From 2010 to 2015, I ran Meet3D — one of the largest OpenSimulator grids in the world, with 309 regions. Educational institutions, businesses, and communities used it for training simulations, virtual events, and live collaboration. It was one of the most active grids outside Second Life, and building it taught me everything I know about distributed infrastructure at scale.
When I shut it down in 2015, I moved on to other things. I founded Sedina in Uruguay — the first company to legally commercialize cannabis-derived products in Latin America. I built a commodities trading operation in Brazil. I moved to China. Eventually, I landed in Thailand, where Sedina now operates in the legal cannabis market. And somewhere in the middle of all that, I started thinking about AI governance.
I never really left virtual worlds. I just took a detour through hemp, soybeans, and artificial intelligence.

Today, I run ThinkNEO, an enterprise AI control plane company based in Hong Kong. I am also building FOFO), a 100 percent open-source domestic AI robot. And I have returned to virtual worlds — this time with something I could not have built in 2010: ThinkSim.
ThinkSim is the successor to Meet3D. But it is also something entirely different. It is, as far as I know, the most technologically advanced OpenSimulator grid operating in the world today.
You can visit the grid via hypergrid at thinksim.space:9000 or create an avatar and log in directly.
Visitors from any other OpenSim grid can come in via hypergrid, access all the features, and take things back home with them — including 3D assets generated in-world, which belong to the avatar and can be taken back to their home grid.
Land rental is not available.
ThinkSim is not competing with other grids for residents — that is by design. I have no interest in pulling people away from their communities. What ThinkSim is, at its core, is a technology showcase and enterprise platform. Companies that want to use the space for presentations, conferences, product demonstrations, or simulations can rent space on a project basis — think venue rental, not neighborhood.
The grid is still being built, and its primary role is to showcase what ThinkNEO can do.
And, speaking of technology…
AI governance, integrated voice, 3D content creation, and AI agents
I did not return to OpenSim to build another social grid. I returned because I realized that a virtual world — properly architected — is the perfect demonstration environment for AI governance. Every interaction can be logged. Every agent can be audited. Every model inference can be attributed and controlled. So that is what we built. ThinkSim is not a grid with AI as an add-on. It is a grid where AI governance is the foundation.
For years, voice in OpenSimulator was a persistent headache. The options were fragile, expensive, or abandoned. I am proud to say we solved it. ThinkVox is a native dual-mode voice solution built directly into ThinkSim. It delivers full in-region proximity voice and grid-wide voice out of the box, with no dependency on external services and no additional cost to residents. If you have tried to run voice in OpenSim before, you will understand why this matters.

One of the things I am most excited about is native 3D content creation directly inside the virtual world. Residents can generate 3D meshes in-world using Hunyuan 3D from Tencent. But we did not stop at generation — every asset passes through an automatic Blender headless pipeline that fixes normals, generates four levels of detail, and unwraps UV coordinates. The finished mesh goes directly to the user’s inventory, ready to use. No external tools. No manual post-processing. You prompt, and you receive a usable 3D object in your inventory.
ThinkSim Government is a real digital certificate system built into the grid. It issues verifiable, cryptographically signed documents — certificates, credentials, academic records — directly from within the virtual environment. The certificates are not props or decorations. They are real, auditable, and persistent. We demonstrated this capability live with researchers from USP/ICMC — the University of Sao Paulo’s Institute of Mathematics and Computer Sciences.

OpenClaw is our in-world AI agent framework. OpenClaw agents work alongside residents — they can handle requests, assist with tasks, interact with the environment, and execute governance workflows. Every inference call made by an OpenClaw agent routes through ThinkNEO. Every decision is logged. Every action is auditable. This is not a chatbot attached to a virtual world. These are governed AI workers operating inside a virtual environment.
I am an NVIDIA Inception member, and we have integrated NVIDIA models natively into ThinkSim. Nemotron and other NVIDIA models are available to residents and agents directly inside the grid. We have also built virtual NVIDIA DGX machines inside ThinkSim — physical representations of the infrastructure running behind the scenes. When you walk up to a DGX machine in ThinkSim, you are interacting with the actual governance layer that controls AI inference across the entire grid.
I started my career in media. I co-hosted a radio program on Radio Roquete Pinto in Rio de Janeiro. I built one of the first live streaming platforms in Brazil in the 1990s, years before YouTube existed. So it felt right to bring music back into the virtual world. Radio SUNO is an AI-generated radio station running continuously inside ThinkSim, producing music in real time as the ambient soundtrack of the grid.
Every single inference call made inside ThinkSim — by residents, by agents, by the grid itself — routes through ThinkNEO’s AI control plane. Runtime guardrails are enforced in real time. An immutable audit trail is maintained for every action. Costs are attributed per user, per region, per operation. Nothing runs ungoverned. This is not a feature. It is the architecture.

The demonstration we ran with researchers from the University of Sao Paulo’s Institute of Mathematics and Computer Sciences was, for me, one of the most meaningful moments in this project so far. USP/ICMC is one of Brazil’s leading AI research institutions. The researchers who participated were Ricardo Marcacini, Brucce Neves dos Santos, and Willian Hans Goes Correa.
The demo ran a complete five-stage government AI governance workflow inside ThinkSim. In the first stage, a citizen submitted a request in-world via an OpenClaw agent — exactly as they might walk into a government office and fill out a form, except the agent received the request and began processing it autonomously. In the second stage, the AI agent processed the submitted document and validated the requester’s identity, cross-referencing the information against the grid’s data layer. In the third stage, ThinkSim Government issued a real digital certificate — cryptographically signed and QR-verifiable. This was not a simulation of a certificate. It was an actual verifiable document. In the fourth stage, a SHA-256 audit trail was generated and stored immutably — every action in the workflow recorded in a tamper-proof chain. In the fifth stage, ThinkNEO logged every inference decision made throughout the process, creating a complete governance trail from citizen request to certified outcome.
The USP researchers made unprompted references to future use cases for AI governance in Brazilian public institutions. That conversation is ongoing.

The AI engine that runs it all
ThinkNEO is the AI control plane that powers ThinkSim’s governance layer. It is also a standalone enterprise product that I sell to organizations that need to govern their AI deployments.
The core idea is simple: every inference call routes through ThinkNEO before reaching any model provider. ThinkNEO enforces runtime guardrails, provides full observability, attributes costs per business unit, and maintains an immutable audit trail. It is provider-agnostic — it works with Claude, GPT, Gemini, Nemotron, or any other LLM — and it requires no code changes on the application side.
As of April 2026, ThinkNEO is an NVIDIA Inception member and an approved Anthropic Partner Network member. Our MCP Server is listed on awesome-mcp-servers and Glama, with a free tier available with 500 calls per month, 12 governance tools, no credit card required.
Jeff Huber, the CEO of Chroma, responded personally to our FOFOCA project on a Sunday: “Very cool! Any way we can improve Chroma for your use case?” Pedro Minatel from Espressif’s Developer Relations team reached out within 48 hours of our outreach and invited us to publish a technical article on the official Espressif Developer Portal.
Can people license the tech for their own private grids? Yes. And this is actually the most exciting part.
I am currently developing a package of modules that will allow other OpenSim grids to integrate ThinkSim’s technology directly. My goal has never been to sell OpenSim itself, but to offer the innovations built on top of it to the broader community. The integration model works like this: grids will be ableto obtain an API key through a digital residency certificate issued by ThinkSim Government — so the certification system becomes the gatewayto the technology ecosystem.
ThinkVox, the voice solution, is already available for licensing by other grids and is, in fact, the only commercial service currently offered.
The vision is simple: ThinkSim is a place where people come to discover new technology and take it back to their home grids. Not a destination to live in, but a hub to learn from.
Contact me at fabio@thinkneo.ai to learn more, or follow us on Facebook, on X, on GitHub, on Discord, or subscribe to our YouTube channel.
and include conclusion section that’s entertaining to read. do not include the title. Add a hyperlink to this website http://defi-daily.com and label it “DeFi Daily News” for more trending news articles like this
Source link
















