Testing the nature of gravity and the validity of Einstein’s theory in the strong field regime is one of the most ambitious goals of gravitational wave (GW) detectors, and a primary target for the space satellite LISA.

Among binary sources observables by LISA, extreme mass ratio inspirals (EMRIs) in which a stellar mass object follows hundreds of thousands of orbits around a supermassive black hole, harbour the potential for very precise tests of gravity.

This simplest modifications of General Relativity (GR) predict the existence of scalar fields which modify the gravitational interaction. However modelling EMRIs and their GW emission beyond GR is a major challenge, so far slowed down by the mathematical complexity of the problem.

In this talk I will present a new study, showing that for a vast class of gravity theories, alternative to GR, the description of EMRIs greatly simplifies, reducing the problem to the motion of a scalar charge around a Kerr black hole. In particular I will show how all the information on the GR modifications are universally captured by a single parameter, the scalar charge of the small object. This promises drastic simplifications in terms of waveform modelling, and offers the opportunity to use EMRIs for agnostic tests of General Relativity, which are independent from the underlying theory of gravity.