When people talk about AI agents and blockchain, most discussions focus on payments. But I think that's only a small part of the picture. AI agents can already make payments through many existing systems — credit cards, stablecoins, traditional banking, and crypto networks. We don't need to reinvent payments just because agents are involved. The harder problems are identity, authorization, verification, proof, ownership, and the exchange of value between autonomous agents. That's where we believe blockchain can play a much more fundamental role. At ArcBlock, we're building blockchain directly into ARC (Agentic Realm Computer) as a foundational infrastructure layer — not simply adding crypto payments to AI applications. And ABT remains the single utility token for the ArcBlock ecosystem. As the ecosystem grows, ABT is designed to support the utility and economic activities that emerge from agents, applications, and services working together. Think of ABT as fuel for the ecosystem — but fuel only creates real value when there's a working engine and something meaningful to power. We're not building AI agents to use crypto. We're building an agentic economy where blockchain and crypto have a natural reason to exist.

AI Cloud Atlas: A Shared Map for People and AI More than just a conceptual map, this is also a roadmap for where we're taking ARC (Agentic Realm Computer). It connects the ideas, architecture, and capabilities we're building into a bigger picture — helping both humans and AI agents understand the world we're creating. A glimpse of what's ahead for ArcBlock, and how the pieces are coming together. arcblock.io/en/articles/atla…

Our new ArcBlock website isn’t built around a traditional CMS. It’s built around an ontology-driven knowledge system. What does that mean? A CMS manages pages and content. An ontology models concepts, meaning, and relationships. Instead of organizing knowledge into isolated pages, we connect ideas, technologies, products, research, and their evolution into a structured, machine-understandable system. With AFS and AFS-UI, the website becomes just one way to explore and interact with that knowledge. Humans can read it. AI agents can understand and work with it. New interfaces can be generated from it. We’re not just building a website that publishes information. We’re building a knowledge system that can continuously evolve. arcblock.io/en/articles/onto…

Over the past 48 hours, we’ve detected several security attacks targeting our infrastructure. Our team immediately investigated and has released security fixes. As a precaution, we’ve patched our online services and rotated their secret keys. So far, we’ve found no evidence of financial loss or compromised user data. However, we did find evidence of attackers attempting to exploit vulnerabilities in older versions of Blocklet Server. If you’re running Blocklets, especially on a self-hosted Blocklet Server, I strongly recommend that you: Update to the latest patched version. Rotate your admin keys and other potentially exposed secrets. Perform a security audit of your deployment. The identified vulnerabilities affect the legacy Blocklet Server 1.x architecture and potentially deployed Blocklets. Our latest ARC (Agentic Realm Computer) architecture is not affected by these specific vulnerabilities. We’ll continue monitoring the situation and sharing updates as needed. Security is never something we take lightly. Thank you to our team for responding quickly and keeping our infrastructure and community safe.

I believe the next bottleneck in software engineering is shifting from writing code to organizing, verifying, and supervising work. At ArcBlock, we call the production model we want to build toward DarcFactory: AI workers operating continuously within clear intent and boundaries, results receiving independent checks, and people able to see what is happening and intervene where judgment is needed. arcblock.io/en/articles/darc…