The analog input's internal resistance can be safely considered so large as to be negligible. The actual leakage current into the input pin will probably be specified in the microcontroller's datasheet. Typical values will be in the \$\mu A\$ or \$nA\$ range and will change slightly when the internal circuitry samples the pin. For most cases, the voltage on either side of R3 will be about the same, so R3's presence can be neglected as well. Assuming all of the current from I1 goes through R1 to ground is valid for most cases.
If, however, R3 becomes sufficiently large, the internal resistance of the analog input can no longer be ignored. This is also true for sufficiently tiny values of I1. But those extremes are rare and can be mitigated with a well designed analog buffer (such as an op amp in voltage-following mode).
One other thing to note is that some microcontroller manufacturers have a limit on the series resistance into a analog input. For example, Microchip specifies a 10\$k\Omega\$ limit on the total resistance going into an analog input pin on their PIC microcontrollers. Your 56\$k\Omega\$ resistor above would cause the ADC function to fail on a PIC.
