Chaos Thinks. Structure Decides.
We don't just publish open problems — we build the community to solve them. Join an independent research collective where we co-design neuro-symbolic experiments, author papers together, and unite to build AI Safety.
Research Projects
CSL-Core Downloads
Accepting Proposals
Research Tracks
Experiment
Run experiments with CSL-Core on real AI agent workflows. LangChain, AutoGen, CrewAI, OpenAI Agents — pick your stack, test our constraints.
Benchmark
Create rigorous benchmarks comparing CSL constraints against traditional LLM guardrails. Datasets, code, and reproducible results.
Paper
Produce arXiv-quality research on formal constraints for AI governance, causal AI, or policy verification.
Integration
Integrate CSL-Core into developer workflows: GitHub Actions, CI/CD pipelines, agent orchestration frameworks.
Open Research Projects
4 projects currently accepting proposals — but more ideas and projects welcome!
CSL Governance for LangChain Agent Workflow
Build a working demo that wraps a LangChain agent with CSL-Core constraints. The agent should fail-safe when constraints are violated, with full causal trace logging.
EXPECTED OUTPUT
- ▹GitHub repo with working demo
- ▹CSL policy file(s) used
- ▹Writeup documenting the integration approach
- ▹Blog post or short paper explaining the results
Have your own research idea?
We don't fund features. We fund ideas. Submit your own proposal.
Submit a ProposalOPEN PROBLEMS IN AI GOVERNANCE
Looking for harder challenges? These unsolved questions define the frontier of our research.
- #01Can formal policy constraints prevent prompt injection attacks?
- #02How can AI agents produce verifiable compliance proofs?
- #03What is the minimal constraint system for safe autonomous agents?
CHIMERA FELLOWS
Your work becomes your credential.
Successful research contributors become Chimera Fellows—recognized members of our research community with co-authorship opportunities and early access to new tools.
Recognition
Listed as a Chimera Fellow on the website and in publications
Portfolio
Published work, GitHub repos, and papers for your research portfolio
Co-authorship
Invited to co-author publications, collaborate on future research, and get early access to tools
Chimera Research Grants
Future research grants will be funded by the Chimera ecosystem. As CSL-Core and Chimera Runtime grow, we will channel resources back into fundamental research. The first funded grants will go to Chimera Fellows.
How to Participate
Three steps. No bureaucracy.
Pick a Problem
Browse open research projects above or check our Open Problems page for unsolved challenges.
Submit Your Proposal
Fill out the proposal form with your approach, timeline, and relevant background. We review on a rolling basis.
Research & Publish
Once accepted, run your experiments. Publish the results openly. Become a Chimera Fellow.
FAQ
Anyone. We don't care about your degree, institution, or pedigree. We care about your ideas, your code, and your commitment to open research. Students, independent researchers, industry engineers — all welcome.
The Chimera Research Program is currently non-monetary. Contributors earn the Chimera Fellow title, website recognition, co-authorship opportunities, and build their research portfolio. Future research grants will be funded by the Chimera ecosystem as it grows.
Yes. All work must be published under an open-source license (MIT, Apache 2.0, or similar). You retain full authorship and intellectual credit. Chimera Lab gets a co-acknowledgment.
Recognition on the Chimera website, co-authorship on related publications, early access to new tools and research, and priority consideration when future funded grants open.
Negative results are results. If your rigorous attempt didn't work, that's still valuable knowledge. We value the effort and the learning, not just the outcome. Document what you tried and what you learned.
We review proposals on a rolling basis. Typical response time is 5-7 business days.
Have a research idea? Submit a proposal or reach out on Discord.