You would need more information to obtain the gound acceleration for a 2% probability of exceedance in 50 yr. What is needed is a seismic hazard curve. This relates the ground acceleration to probabilities. The relationship is not linear due to many things such as seismicity and ground attenuation relationships.
I'm not sure if you're actually looking for Santiago's ground acceleration, or if you just want to know how to convert accelerations due to varying probabilities. But, for what it's worth, the US Department of Defense has the following ground accelerations for Santiago for 2% exceedance in 50 yrs.:
Ss=1.65g (ground acc. for buildings w/period of 0.2s)
S1=0.75g (ground acc. for buildings w/period of 1.0s)
These values are probably very approximate and conservative, though.
Hope this helps you out a little.