Excited that zkAPI is out and live! Great joint effort with @VitalikButerin on our initial design, @curryrasul and @kenziyuliu + OA team on implementation, and the amazing push from @VittoStack on the dAI team to bring this out. If you want to implement private payments on Ethereum for your API, check the post and repo. We have more things coming out to enhance user sovereignty and privacy in the age of AI.

dAI <> @MiniMax_AI one of the leading teams in open source AI hosted us in Shanghai we discussed the infra we’re building for the open agent economy, multiagent AI, security grateful to meet head of eng @SkylerMiao7 - he shared impressive work on agent orchestration & feedback

dAI Team in China this week! We’re meeting so many people that are excited about Open Source AI and the properties that Ethereum can deliver. Scaling multi-agent systems, making AI more private, secure, and servicing the human users.

1/Introducing ACTA: Anonymous Credentials for Trustless Agents. A composable privacy layer above ERC-8004 so agents can prove: personhood, reputation, model provenance, user jurisdiction, and more — without publishing the interaction graph. 🧵

We’re proud to team up with @ethereumfndn’s dAI team to offer strategic guidance to AI projects building on Ethereum & X Layer. We'll provide infrastructure, ecosystem support, and collaborate on initiatives that shape the agentic economy. Stay tuned for more updates.

Sharing a super simple, user-owned memory module we've been playing around: nanomem The basic idea is to treat memory as a pure intelligence problem: ingestion, structuring, and (selective) retrieval are all just LLM calls & agent loops on a on-device markdown file tree. Each file lists a set of facts w/ metadata (timestamp, confidence, source, etc.); no embeddings/RAG/training of any kind. For example: - `nanomem add ` starts an agent loop to walk the tree, read relevant files, and edit. - `nanomem retrieve ` walks the tree and returns a single summary string (possibly assembled from many subtrees) related to the query. What’s nice about this approach is that the memory system is, by construction: 1. partitionable (human/agents can easily separate `hobbies/snowboard.md` from `tax/residency.md` for data minimization + relevance) 2. portable and user-owned (it’s just text files) 3. interpretable (you know exactly what’s written and you can manually edit) 4. forward-compatible (future models can read memory files just the same, and memory quality/speed improves as models get better) 5. modularized (you can optimize ingestion/retrieval/compaction prompts separately) Privacy & utility. I'm most excited about the ability to partition + selectively disclose memory at inference-time. Selective disclosure helps with both privacy (principle of least privilege & “need-to-know”) and utility (as too much context for a query can harm answer quality). Composability. An inference-time memory module means: (1) you can run such a module with confidential inference (LLMs on TEEs) for provable privacy, and (2) you can selectively disclose context over unlinkable inference of remote models (demo below). We built nanomem as part of the Open Anonymity project ( but it’s meant to be a standalone module for humans and agents (e.g., you can write a SKILL for using the CLI tool). Still polishing the rough edges! - GitHub (MIT): - Blog: - Beta implementation in chat client soon: Work done with amazing project co-leads @amelia_kuang @cocozxu @erikchi !!

Terence Tao proposes what he calls a "Copernican view of intelligence". Instead of buying into the common, one-dimensional narrative that artificial intelligence will simply evolve from "subhuman" to "superhuman" and ultimately make humanity entirely redundant, Tao urges us to look at the bigger picture. Much like the Copernican revolution proved the Earth is not the center of the universe, Tao suggests we need to realize that human intelligence isn't the only, or necessarily the highest, form of intellect. Historically, we have treated other forms of storing or creating knowledge—like animals, books, and computers—as secondary. However, we actually exist within a much richer universe of intelligence. Both human intelligence and computer intelligence possess their own distinct strengths and weaknesses. The true potential lies not in viewing them as direct competitors, but rather in focusing on collaboration. By working together, humans and computers can achieve additional things that neither could accomplish on their own, requiring us to think in much wider terms than just what humans or computers can do alone.

Excited to present ERC-8211, an effort led by @biconomy and supported by our Improve UX track! Smart batching fills a gap between users, agents and complex onchain execution, including crosschain execution. --- In our interop work, we've identified three layers of functionalities: - Representation: A crosschain action is made machine-legible via some structure, eg., an array of calls à la ERC-5792 or an intent ("I want X"). - Orchestration: Particularly in asynchronous settings, actions are composed with multiple steps. Some entity is required to trigger these calls at the right time, in the right place and with the right inputs. - Liquidity/message-passing: Moving liquidity crosschain is the goal of intent-based bridges. More generally, passing messages, beyond just value, is the critical building block of crosschain actions. We've been working on the liquidity layer with Open Intents Framework. The framework provides standard, secure and modular components towards instantiating a common marketplace for fast liquidity. Meanwhile, orchestration may be delegated away, with another party taking on the duty to solve an intent or triggering the right calls. However, with the rise of personal agents, it is much more viable for the user to represent and control the orchestration of their complex crosschain actions, via their agent. --- Enters ERC-8211. Smart batching represents execution to make it: - Sufficiently concise for user-friendliness, unlocking the excellent UX of a single signature for many operations - Sufficiently flexible for agent-friendliness, with dynamic values obtained during runtime, and predicates gating execution as it happens, for further user security Smart batching is also friendly to composition with different accounts and different bridge models. This means using secure account and bridge primitives, safety is preserved even as the agent vibes its way through bigger smart batches with more complex steps. --- This is an invitation then to build new ways of interacting onchain and crosschain, with agents as guides and smart batching under the hood. Find more details in the linked post and on

11 months ago we were hacking on a random idea about moving money on the internet. Now we have legendary companies and industry leaders supporting this open protocol. This is a huge win for open standards. Ever since I got addicted to stumbleupon back in my youth, I’ve known open = better. There’s no better way to build a tech that lasts, adapts and wins.

Excited to launch Gemma 4: the best open models in the world for their respective sizes. Available in 4 sizes that can be fine-tuned for your specific task: 31B dense for great raw performance, 26B MoE for low latency, and effective 2B & 4B for edge device use - happy building!

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So: Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them. IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides). Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale. Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base. Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into. Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries. Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows. TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

Today is 4.02 day. Less than a year ago (May 6, 2025), we published the x402 whitepaper. Since then, the protocol has: • Processed 166M+ transactions • Unlocked new monetization models for API, data, and other providers on the web. • Enabled autonomous flows that were difficult to support with traditional account infrastructure Today, we’re taking the next step. x402 is moving to @linuxfoundation, and we are launching x402 Foundation - with broad, cross-industry collaboration & input to steward an open, neutral standard for payments over HTTP. A big thank you to our amazing founding partners in this effort: @Adyen, @Awscloud , @AmericanExpress, @Ampersend_ai, @Circle, @Cloudflare, @Coinbase, @Fiserv, @Google, @KakaoPayCorp, @Mastercard, @merit_systems, @Microsoft, @Polygon, @PPRO_Payments, @Shopify, @SierraPlatform, @Solana , @Stripe, @Thirdweb, and @Visa. Enrollment for the foundation is open if you - or your company - is interested in participating, please reach out!

New public experiment: Model Fusion Use multiple models, analyze outputs, and fuse the results for a response that every Deep Research agent preferred to its own, in our testing. No subscription needed at all.

ERC-8004 is now live on mainnet. 5 months ago, we wrote the specs for the Trustless Agents standard. Since then, over 10k agents registered on testnet. Today, we’re releasing it on Ethereum Mainnet. Welcome to the 8004 Genesis Month. Here’s everything you need to know 👇