Especially since the balance of the evidence you've linked to appears to support that assumption.
It's not as if "there is at least some level of marijuana use incompatible with driving" is exactly an extraordinary claim; the burden of proof for establishing parameters for a "safe" level of high whilst in control of a vehicle really ought to sit with proponents of smoking and driving.
Put another way, jumping to the conclusion that driving under the influence of marijuana isn't safe isn't likely to kill people if it's incorrect.
Defaulting to the position of criminalization regardless of the evidence, particularly in this context given the history and impact of existing marijuana policy, will likely perpetuate the creation of flawed policy.
I, for one, think we should be open to actually studying the topic instead of baselessly and unscientifically concluding that driving under the influence of marijuana is equally as dangerous as driving under the influence of alcohol.
The research about Marijuana's impact on driving is inconclusive, but there are indications that it might have no negative impact (for the most recent example: https://s3.amazonaws.com/s3.documentcloud.org/documents/1667...) and even that it might reduce risks of crashes.
As a result, it's unfortunate that it's apparently socially acceptable to assume without question that marijuana has a negative impact on driving. New research could still demonstrate that it does not, just as new research has been strongly suggesting that our other assumptions are likely flawed. If marijuana is found to not be dangerous for drivers, then it will very unfortunate that we've promoted yet another misconception about it due to unchecked assumptions.