Testing & Evidence
AI Internet-Meritocracy (AIIM) is an experimental system. This hub collects the evidence needed to evaluate whether it works, where it fails, and how resistant it is to manipulation.
Independent reviewers wanted
Are you a researcher, engineer, nonprofit or governance specialist, AI-safety researcher, economist, open-science researcher, security researcher, or another relevant expert?
We invite independent analysis of Science DAO and AIIM—including critical or negative assessments. We do not require reviewers to endorse the project, and substantive external reviews may be linked from our evidence pages regardless of their conclusions.
What we are testing
- Agreement between AIIM evaluations and independent human judgments
- Stability across repeated evaluations and different models
- Resistance to prompt gaming, strategic self-presentation, and adversarial manipulation
- Reproducibility of evaluation procedures
- False-positive and false-negative allocation decisions
- Appeals, edge cases, and known limitations
Adversarial testing
Our adversarial testing program is designed to expose failure modes rather than hide them.
Read the adversarial testing plan →
Related methodology
Why AI science funding needs adversarial testing →
Preventing the prompt-gaming problem →
Current product status and limitations →
Independent review policy and reviewer packet →
Independent review
Independent scrutiny is part of the evidence process, not a testimonial program. Reviewers retain control of their conclusions and are encouraged to publish on venues Science DAO does not control.
Review Science DAO / AIIM independently →
Publication principle
Results should be reported whether they support or challenge AIIM. As experiments are completed, this page will link to preregistrations, datasets, evaluation protocols, independent reviews, and result reports.
Current priority: produce a small, reproducible public validation experiment before attempting large-scale deployment.
Support Independent Science
Our flagship product, AI Internet-Meritocracy, is an app (in the stage of open beta-testing) designed to allocate donated funds to researchers and open-source developers using AI-assisted evaluation of documented contributions. Payments depend on available funds and eligibility requirements.
Help fund the proposed five-month public test of AIIM’s allocation model with distribution of $1000 to real salary recipients. Support the next testing milestone.
Supporting independent science is not only a matter of fairness to researchers whose expertise and work are often underfunded. It is also essential for addressing systemic failures in scientific publishing that delay discoveries and leave important results unnoticed. In science and software, even one missing component can prevent an entire system from working.
Help valuable research and open-source infrastructure move forward. Please make a donation to support independent scientists and free software developers.
Independent review
Researchers and AI-safety reviewers: We invite independent technical criticism of Science DAO, AIIM, and related AI-safety work, including critical conclusions. Please examine assumptions, human-in-the-loop mechanisms, prompt-injection risks, governance, and failure modes. Use the independent-review packet →
Disclaimer
Experimental-system notice: AI Internet-Meritocracy is an experimental funding system. Its AI-generated evaluations are heuristic judgments based on available public or connected-account evidence; they are not validated measurements of a person’s causal economic or scientific impact. Payment transactions are already recorded on-chain and can be verified on the blockchain. The current beta initiates payments off-chain through Node.js and uses custodial and administrative components. Decentralized governance and non-custodial wallets remain under development; on-chain payment records are already available. Evaluations may contain factual errors or biases and should be interpreted together with audit logs, appeals, human oversight, and published test results.