Well ... the AK103 uses 7.62 mm bullets , while the LR 300 uses 5.56 mm bullets .
So the LR 300 is much more accurate than the AK103 while the AK103 makes much more damages to enemies ... so the rifle accuracy is quite inversely proportional to the bullets dimension : much power = less accuracy and viceversa.
There's another famous assault rifle using 5.56 mm bullets : the M16.
It uses such small and penetrating bullets because its main target is not to kill enemies , but to stop them ( with one or more M16 bullets holes or with 2 or 3 bullets inside the body , the enemy is probably not letally vounded but for sure simply can't move ).
AK103 bullets in terms of damage given to enemy are much more letal than LR 300 ones but at mid or high range the rifle absolutely has not an acceptable efficacy if used in auto mode. ( U can try by yoursefl .... in the game of course
In auto mode it has a significant recoil that makes its control and its accuracy acceptable only at low range.
At mid and high range u can control its aim ( but not so easily ) only at burst or semi mode.
The problem I've tried to spot is that if u see a player using AK103 at high/mid range in auto mode and with a high accuracy , he/she is cheating with a probability > 90 per cent .
U can see infos about the weapon used when u are vounded/killed so u can control the sequence of the attack and if the player has a score of 21 kills - 1 death , that means that he/she doesn't come so often - never ? - near the enemy and as a consequence he/she has an impossible - fake - aim.
If u hear also the AK103 sound and u clearly can hear that it's used in auto mode , for sure it is a case of cheating.