Ironically, the obverse task -- trying to create a molecule that acts on similar neurobiological targets as an illegal drug but doesnt get people high or look like a molecule from an illegal drug -- is similarly fraught. A flagrant example is BIA 10-2474, an experimental drug meant to target the endocannabinoid system in a roundabout way -- not by directly activating receptors, but by inhibiting an enzyme that degrades endogenous chemicals that activate endocannabinoid receptors -- which would have similar effects as directly activating those receptors. A trial of it killed one person and irreversibly neurologically damaged a few others.
We criminalise drugs at deep costs to research and medecine (I haven't even gotten into the use of MDMA in psychotherapy or ketamine for depression), and regardless on what side of the law they work on, it is a very difficult task for chemists to try to imitate drugs that have been with humanity for millenia when they aren't allowed to make molecules that look like blacklisted ones.
Now, if it was legal people wouldn't have to resort getting that crap as a replacement.
Can you imagine a THC analog that's 10^4 times more potent? Is overdose really impossible? I suspect that we'll find out, in a few years :(