David Nutt shares your hypothesis and is working on such a substance [1]. I agree with you that something like chronic Diazepam abuse is probably less harmful

David Nutt shares your hypothesis and is working on such a substance [1]. I agree with you that something like chronic Diazepam abuse is probably less harmful to the body than a similar alcohol habit - but I'm skeptical (but hopeful) one could design an effective, recreationally-useful GABA-ergic drug that didn't come with the baseline shift and rebound anxiety (and in the limit, seizure/death) usually associated. You can already choose to abuse pills in contrast to alcohol, and it's generally not very pretty either. [1] https://www.theguardian.com/science/2019/mar/26/an-innocent-drink-could-alcosynth-provide-all-the-joy-of-booze-without-the-dangers https://www.theguardian.com/science/2019/mar/26/an-innocent-...