 Leibniz, dreamt of artificial intelligence in 1685. He dreamt of a language so complete that all of human thought could be represented in it. And so precise that if you had a disagreement between any two humans, you could summon a human or a machine and tell it, Calculemus, let us compute and you would have a solution. Now we wanted to understand arguments and reasoning in science and resolve disagreements between scientists. David Hilbert gave us an ideal model for this. He said that the best reasoning in science should be axiomatic. You start with a small set of axioms, statements that you assume to be true and you reason with the axioms and the axioms alone towards your goal so that the validity of your statements can be reduced to the validity of your axiom set. The sixth of his famous 23 problems was to find the axioms of physics. Now this is what we did. We considered Einstein's special theory of relativity and we looked for a small set of axioms following Hilbert that could be written in a computable precise language that our AI could understand following Leibniz. Working with collaborators in Hungary, we found them. We found six axioms of the special theory of relativity written in precise mathematical logic. Now here's what we did with them. We built an artificial intelligence and you can find a demonstration on YouTube which will take steps in your physics proof and verify each step as being right or wrong and we think we can go much further than that. We want to build a sort of Siri or Alexa for scientists that will sit with you as you write your theoretical physics proof and it will suggest steps for you or fill in gaps in your reasoning. We call our paradigm for this goal computational axiomatic science. It is a bridge between the artificial intelligence of today and a 400-year-old dream. This is research with a fascinating history and a grand vision for the future. Thank you for listening.