They can.
Some years ago there was a spell of around 3 weeks where the bottom two jackpots hit far less frequently and grew to levels much higher than usual between hits. The Minor routinely rose to 1000, and often 2000 and 3000 between hits, and the Mini tended to hit past 100. Microgaming would never comment nor explain even though it was so bleedin' obvious to anyone playing, and at the end of these few weeks things were restored to what we now regard as normal.
The suggestion was that it was a trial to see whether less frequent but more substantial lower jackpots was a better lure for players than the "pointless" tenner and ton we get now. If they can tweak the hit probability settings for a 3 week trial, they can do so for another trial, or indeed to produce bigger and less frequent major jackpots.
Of course, it does not prove they have actually done it, we would need to see a number of Mega jackpots building to 10M+ from now in contrast to none having done so previously, then a statistical "proof" could be calculated as to whether it's down to an actual change, or just as likely to be down to random fluctuations. Statistical "proof" is usually regarded to have been demonstrated when it reaches the 95% confidence level, as in 95%+ probability that the result is down to a change in the hit algorithm and <5% that it's just a freak random fluctuation.
The very first jackpot hung on past 5Million, and there was talk then that MGS had deliberately manipulated the first hit to be so high in order to generate PR for the launch and first major payout for the new game. Since then though, it has often soared high as well as hit much lower down. One 10M+ in all that time is more than likely just an unusual outlier event.
MGS could have helped themselves by not covering up what happened for those 3 weeks where the lower jackpots kept getting so high, because there is no way on earth these can statistically be shown to be part of the same "system" as all the other data for these jackpots.