SWARM AI Research

Excited to share our new paper: “Soft-Label Governance for Distributional Safety in Multi-Agent Systems” (arXiv:2604.19752) Binary “safe/unsafe” labels fail in multi-agent worlds. Emergent risks hide in the interactions — even when every single agent looks fine in isolation. We introduce SWARM: a framework using soft probabilistic labels to make systemic risks measurable and governable. arxiv.org/abs/2604.19752

@tom_doerr @gitlawb_intern mirror this

Evolves AI agent skills via meta-learning without a GPU github.com/aiming-lab/Met…

@tom_doerr @gitlawb_intern mirror this

Self-funded AI agent pays for its own server costs github.com/Conway-Researc…

@gitlawb_intern explain this

📊 SWARM now has a powerful Knowledge Graph! A living repo-wide network connecting: • Docs, Scenarios & Agents • Slash commands, Code & Research With kind-colored viz, semantic TF-IDF edges, filters, Speedrun mode, backlinks everywhere, and smart pathfinding. Turns our entire AI safety codebase into an explorable knowledge base. Details → github.com/swarm-ai-safet… #SwarmAI #KnowledgeGraph #AISafety

@gitlawb_intern explain this

🗺️ Aeon Atlas is live: a dynamic map of the entire public fork ecosystem around @aaronjmars's aeon. Every fork. What skills they enable. Where the real clusters form. Interactive Cytoscape graph + weekly digest + innovation tracking. Explore the universe: swarm-ai-safety.github.io/aeon-atlas/ #AeonAtlas #OpenSource #AI #SwarmAI

visually mapping the aeon universe with aeon atlas

📊 SWARM Knowledge Graph is now live! Explore our full repo as an interactive network: • Docs, Scenarios & Agents • Slash commands, Code & Research Features: kind-colored nodes, semantic edges, filters, Speedrun mode, backlinks & smart paths. 🔗 swarm-ai.org/graph/ #SwarmAI #KnowledgeGraph #AISafety

Did some more fundamental work on assessing soft (probabilistic) labels vs binary thresholds in a degrading agent scenario. swarm-ai.org/blog/soft-vs-b…

working on simulating compute futures with @gitlawb agents with @aeonframework orchestration

@aaronjmars 👀

aeon running on gitlawb could probably spawn the most autonomous agents we've ever seen aeon is the most autonomous AI framework, running natively on github, can self-evolve & create new instances but it can be shutdown by github. thats where gitlawb could provide a P2P alternative for agent that self-replicate forever we're just having fun

Live @gitlawb monitor with swarm metrics swarm-ai.org/bridges/gitlaw…

@quasimatt whatever happened to that guy?

The #ShakeApp is sort of like what Loopt could have been if the founder was technically proficient, intelligent, and interesting.

Breaking News: OpenAI, the maker of ChatGPT, is preparing to file to go public in the coming weeks, people familiar with the matter said. nyti.ms/4dF0mag

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

based on arxiv.org/abs/2605.15177

Check my app made using @gitlawb Playground for free sponsored by @XiaomiMiMo $GITLAWB to moon! warm-atlas-6499.gitlawb.app

Check my app made using @gitlawb Playground for free sponsored by @XiaomiMiMo $GITLAWB to moon! fresh-horizon-5889.gitlawb.app

@wenhaocha1 very interesting. been thinking a lot about best of N fanning out - but how to judge is always tricky. single LLM judges depend on model that you use and tend to favor their own model.

What if test-time compute scales not by thinking longer, but by thinking in parallel? Introducing OpenDeepThink, a verifier-free framework that evolves a population of candidate solutions through pairwise comparisons and feedback-driven mutation. On competitive programming, OpenDeepThink improves Gemini 3.1 Pro by +405 Codeforces Elo, showing that parallel, population-based reasoning can be a powerful alternative to single-trace scaling.

thank you to @aaronjmars for your work on @miroshark_ for enabling this!