An open research organization
Mathematics
for AI safety
We state open problems precisely, and check their solutions in public.
Good mathematicians prove theorems, great mathematicians come up with conjectures, and the greatest mathematicians come up with definitions.
Asking the right questions
Mathematicians prove theorems, ask questions, build theory and apply it. MAIS works on the second: stating the questions AI safety needs answered.
Proving theorems
Turning a conjecture into certainty.
Asking questions
Finding the question worth answering, and stating it precisely. This is where MAIS works.
Theory building
Connecting results into a framework that explains them.
Applying theory
Using that framework to build and check real systems.
Where to begin
One starter problem per agenda, each sized for a first paper.
The map
92 problems in eight agendas, drawn as a survey chart. Find where your mathematics fits.
Open the map ↗