Evaluation Foundation
Building the profession of independent AI evaluation.
AI should serve the people whose lives it shapes. We’re building the community, capabilities, and shared infrastructure to help make that possible.
Explore our workWho gets to decide
what “good” means?
Beyond the builder’s view.
When the companies building AI also define what “good” means, their tests can miss the experiences of the people who live with the results.
Experience is expertise.
Domain experts and regular people bring knowledge that benchmarks alone cannot capture. They need a meaningful role in deciding what gets evaluated.
Better evidence leads to better decisions.
Independent evaluation asks not just whether a system works, but for whom, in which contexts, and at what cost. It gives people a stronger basis for deciding what belongs in their lives.
A profession.
A community.
A public good.
The Independent AI Evaluation Foundation exists to catalyze and professionalize the community of AI evaluators working where the stakes for society are highest.
We bring together evaluators, domain experts, affected communities, policymakers, and the organizations building and using AI. We tackle fields in which investment in evaluation has not kept pace with technological change.
Our goal is a credible, economically sustainable field of independent evaluation, with the skills, tools, and standards to help technology serve people and institutions well.
Read our mission & visionThe conditions for independence.
A durable evaluation profession needs more than good intentions. It needs opportunity, expertise, shared resources, and the technology to do rigorous work.
Build with usFunding & projects
Access to funding and meaningful projects that create and sustain an economically viable community of professional evaluators.
Skills development
Training and upskilling for evaluators, domain practitioners, and organizations seeking independent evaluations.
Market commons
A shared place to access tools, workflows, best practices, job opportunities, and expert evaluator talent.
Technology infrastructure
Shared evaluation capabilities that lower costs and raise rigor, with primary technical infrastructure provided by Humane Intelligence PBC.
Where we start:
education.
AI is entering the lives of students, teachers, and families. We’re starting here because how we evaluate these systems will shape not only what people learn, but who they have the opportunity to become. We start with vulnerable and marginalized students.
Dignity & agency
Human
centricity
Does AI preserve dignity, agency, and the primacy of human relationships in learning?
Reasoning & creativity
Cognitive
flourishing
Does AI support reasoning, creativity, and durable understanding rather than substitute for them?
Opportunity & futures
Economic
flourishing
Does AI in education expand, rather than narrow, learners’ long-term economic opportunity?
A field is built together
Bring your expertise.
Help shape the field.
Whether you evaluate AI, work in education, represent a community, or want to support independent evidence, there’s a conversation to start.
Get in touch