Ep 16: Cannabis in the Mail, and the FDA on TikTok — The Shift

Regulators are listening to Reddit, the Postal Service may deliver medical cannabis in Georgia, and Cambridge adds a piece to the health puzzle.

In this episode

• The FDA is seeking an outside contractor to monitor cannabis and hemp conversations on TikTok, Instagram, and Reddit. Per Marijuana Moment, the monitoring would capture products, ingredients, claims, consumer conditions, and sentiment, not just mention volume. The agency also wants a survey of actual dispensary customers covering product preferences, desired effects, medical conditions, frequency of use, average purchases, and preferred dosages.
• The operator read on that: regulators have clinical studies, lab testing, and adverse event reporting, but none of it shows how millions of people talk about cannabis outside a controlled setting. The useful distinction in the episode is between using social listening to identify a question and using it to answer one. Popularity does not turn an anecdote into evidence, but it does tell regulators what to ask about next.
• Georgia's Medical Cannabis Commission voted on August 13 to propose allowing medical cannabis delivery through the United States Postal Service plus FedEx, UPS, and DHL, per Cannabis Business Times. The state has more than 45,000 registered patients, and executive director Andrew Turnage frames it as an access problem in a geographically large state. A public hearing is set for September 16.
• The proposed rule is narrower than the headline: Georgia dispensaries shipping to registered Georgia patients inside Georgia, with certified mail or equivalent, tracking, identity and patient registry verification, and a signature from the authorized recipient. Interstate commerce is a separate fight. The larger question the episode raises is when cannabis stops building a parallel system and starts using the infrastructure medicine already has.
• University of Cambridge researchers combined 32 studies covering more than 100 million people and found cannabis use associated with roughly a 37% higher risk of stroke, with genetic liability to cannabis use disorder linked to increased stroke risk, particularly large artery stroke. The researchers noted substantial differences between studies and limitations in the observational evidence, and prior cardiovascular research has not been consistent.
• The closing argument is about maturity rather than the study itself. Cannabis has been stuck between uniquely dangerous and harmless because it is a plant, when the industry already handles that tension fine for alcohol, caffeine, and prescription drugs. Real normalization looks like cannabis being boring enough to discuss honestly: benefits where evidence supports them, risks where it supports those, and uncertainty where the science is unfinished.

Chapters

• 0:09 — The FDA wants to listen
• 0:46 — What the monitoring covers
• 1:25 — A dispensary customer survey
• 2:02 — What regulators normally see
• 2:21 — Social media is messy
• 2:57 — Filling gaps in understanding
• 3:17 — Survey vs. observation
• 3:54 — Messiness can be useful
• 4:31 — Observable human behavior
• 5:08 — Which parts to believe
• 5:45 — Questions, not answers
• 6:22 — Signal vs. chaos
• 6:41 — Cannabis in the mail
• 7:19 — Georgia's USPS proposal
• 7:37 — 45,000 registered patients
• 7:55 — Still just a proposal
• 8:32 — Access, not revolution
• 8:50 — Why drive several hours
• 9:08 — Treating it like medicine
• 9:45 — The other side of the case
• 10:05 — Verification and tracking
• 10:25 — In-state shipping only
• 10:45 — Mundane changes matter
• 11:22 — Using existing infrastructure
• 11:41 — When it stops being remarkable
• 12:00 — News we may not want
• 12:37 — The Cambridge stroke study
• 13:16 — What 37% actually means
• 13:54 — Mendelian randomization
• 14:13 — Cannabis use disorder link
• 14:31 — Science is unsatisfying
• 14:50 — Conflicting prior research
• 15:26 — Two ridiculous extremes
• 16:04 — Alcohol, caffeine, and risk
• 16:23 — Boring enough to be honest
• 16:43 — Questions worth asking
• 17:22 — A more honest conversation
• 17:41 — Demanding better evidence

Transcript

Auto-generated from the episode audio and lightly edited for readability.

0:09 Apparently, the FDA wants to know what you're saying about cannabis on TikTok and Instagram and Reddit. Yes, the same FDA we normally associate with clinical trials, pharmaceutical uh approvals, laboratory data, and very serious people

0:27 reading very serious reports. now wants to listen to social media. And when I first read that headline, I'll admit I found it slightly amusing because if you're looking for reliable scientific information, Reddit would not necessarily be the first place I'd sent

0:46 you. But then I looked into what the FDA is actually asking for. And this became much more interesting. According to Marijuana Moment, the FDA is currently seeking an outside contractor to monitor conversations about marijuana and hemp products across platforms, including Tik

1:06 Tok, Instagram, and Reddit. And they're not simply looking for how many times somebody says the word cannabis. The proposed monitoring would capture information about products, ingredients, claims, consumer condition, sentiment, and other

1:25 characteristics of what people are discussing online. But here's where it gets even more interesting. The FDA also wants research involving actual dispensary customers. According to the same reporting, the agency wants a survey examining things like product

1:43 preferences, desired effects, medical conditions, frequency, and history of use, average purchases, and even preferred dosages. Now, that is fascinating because think about the information regulators normally have available to them. clinical studies,

2:02 medical reports, laboratory testing, adverse event reporting, controlled research, all incredibly important, but none of those necessarily tell you how millions of people are actually talking about cannabis when nobody in a white coat is standing in the room. Social

2:21 media does. And yes, obviously, social media also is full of absolute nonsense. People repeat things they heard from somebody who heard it from somebody else. An algorithm can take one ridiculous health claim and put it in

2:39 front of 10 million people before breakfast, literally. So, no, TikTok comments should probably not replace clinical trials anytime soon, but that's not really what the FDA appears to be trying to do. The agency says its

2:57 cannabis work involves developing cross agency strategy and policy around cannabis products. And this kind of social listening can potentially fill gaps in understanding how products are actually being discussed and used outside controlled settings.

3:17 And that raises a question I I genuinely don't know the answer to. What is more valuable when you're trying to understand realworld customer behavior? Asking a few thousand people carefully designed questions

3:35 or quietly observing millions of conversations people were already having because those two methods can give you very very different pictures. In a survey, people know they're being studied on Reddit.

3:54 Trust me, they usually behave like nobody's studying them at all. I mean, and sometimes that messiness is useful. Somebody might mention an unexpected effect, a product being used in a way regulators didn't anticipate, a dosage trend, a health claim suddenly spreading

4:13 across TikTok, or a new cannabinoid becoming popular before traditional data sets even notice it. That's information. Not necessarily good information, not necessarily true information, but information nonetheless.

4:31 And this is where I think the story becomes much bigger than cannabis because governments are beginning to operate in a world where enormous amounts of real time human behavior are publicly observable. We voluntarily document what we buy, what we take, how it makes us feel, what we believe.

4:50 what we're afraid of and what somebody on TikTok convinced us to try last Tuesday. Researchers 20 years ago would have dreamed of having access to that kind of data set. Today, the problem isn't finding the

5:08 information. It's figuring out which parts of it deserve to be believed. And that's the part I'd be watching because there is a very big difference between using social media to identify a question and using social media to answer one. If thousands of people suddenly start reporting the same

5:27 unexpected experience, perhaps that's something worth investigating. But popularity doesn't turn an anecdote into scientific evidence. 10 million views don't make a medical claim. So maybe the real value of social media isn't that it tells regulators what to

5:45 believe. Maybe it tells them what they should be asking about next. And that's really the shift. The FDA listening to cannabis conversations online sounds funny until you realize that TikTok, Instagram, Reddit, and platforms like them have become one of the largest

6:04 continuously updating records of consumer behavior we've ever created. The question isn't whether there's useful information buried inside all that noise. There almost certainly is. The question is whether anyone can reliably separate the signal from the

6:22 absolute chaos. Here's a sentence I genuinely didn't expect to say on this podcast. The United States Postal Service may soon be delivering cannabis. Yes, the same people bringing you your electricity bill, birthday cards, and the package you forgot you ordered three

6:41 weeks ago could potentially be delivering medical cannabis directly to patients homes. And no, cannabis has not suddenly become federally legal across America. We know that like that's no news. This is happening in Georgia and

7:00 the details are actually much more interesting than the headline. So, according to Cannabis Business Times, the Georgia Medical Cannabis Commission voted on August 13th to propose a rule allowing medical cannabis to be delivered through the United States Postal Service,

7:19 as well as private carriers, including FedEx, UPS, and DHL. Georgia currently has more than 45,000 registered medical cannabis patients and the proposal would allow licensed dispensaries to ship directly to

7:37 registered patients, caregivers, and certain health care institutions within the state. Now, before anyone starts waiting beside their mailbox, this is still a proposal. The rule is open for public

7:55 comment and Georgia regulators are expected to consider it at a public hearing on September 16th. But I think the fact that we're even having this conversation is remarkable because think about how bizarre the relationship between cannabis and the federal government has

8:13 been historically. And now we're discussing whether a federal agency, the United States Postal Service, could become part of the infrastructure delivering medical cannabis to patients. That's quite a sentence. But here's where I think this becomes genuinely interesting. Georgia regulators aren't

8:32 presenting this as some revolutionary new cannabis experience. They're talking about access. Andrew Turnage, the executive director of Georgia's Medical Cannabis Commission, explained that Georgia is geographically large and that patients have struggled to access the medication

8:50 the state regulates. His argument is essentially very simple. Why make a registered patient drive several hours if their medication can safely come to them? And suddenly the question becomes uncomfortable in a very interesting way.

9:08 If we're going to call cannabis medicine, should we also start treating it like medicine? Because Americans already receive prescription medications through the mail? That's completely normal. Nobody sees a pharmaceutical package arrive and thinks, "My god, the postal service has entered the drug

9:26 trade." We understand the distinction. It's medication. It's regulated. it's going to a patient. So if medical cannabis is legally recognized within a regulated medical system, why should the delivery mechanism automatically be different? But there is another side to this. Cannabis isn't an ordinary

9:45 prescription medication under federal law. And putting an intoxicating controlled substance into the mail obviously creates questions. Who receives the package? How do you verify them? What happens if it's delivered to the wrong address? How do you prevent diversion? Georgia's proposal tries to answer some of those

10:05 questions. Reporting on the proposed rule says shipments would require certified mail or an equivalent service, tracking, identity, and patient registry verification and a signature from the authorized recipient. And there is another extremely important limitation. We're not talking about

10:25 somebody in Georgia ordering cannabis from California. The proposal is specifically about Georgia dispensaries shipping to registered Georgia patients inside Georgia itself. Interstate commerce is an entirely different conversation. But this is why I find the story so interesting. Cannabis policy

10:45 usually changes through enormous announcements, legalization, rescheduling, new legislation, but sometimes the more important changes are incredibly mundane. Can it go in the mail? Can a pharmacy dispense it? Can a doctor prescribe it? Can insurance

11:03 eventually cover it? Those questions aren't particularly glamorous, but they're the questions that determine whether something actually functions like medicine in everyday life. So, here's a question I would leave you with. If cannabis is going to occupy a legitimate place in medicine,

11:22 at what point should we stop building an entirely separate system around it? And at what point should it simply start using the infrastructure that medicine already has? I don't know exactly where that line should be, but Georgia may be about to test it. And that's really the shift.

11:41 The remarkable part isn't that cannabis could arrive in someone's mailbox. The remarkable part is that we're reaching a point where for a registered patient receiving regulated medication, that idea might eventually stop sounding remarkable at all. I want to finish

12:00 today's episode with something slightly uncomfortable because on this show, we've talked about cannabis becoming medicine. We've talked about cannabis beverages. We've talked about changing consumer attitudes. We've talked about access. We've talked about normalization. So, I think it's only

12:19 fair that when research comes out suggesting something we may not particularly want to hear, we talk about that, too. And this one definitely got my attention. So researchers from the University of Cambridge recently conducted one of the largest anal an

12:37 analysis ever looking at recreational drug use and stroke risk. According to the University of Cambridge, the researchers combined evidence from 32 previous studies involving more than 100 million people and among the substances they examined

12:56 was cannabis. The result, cannabis use was associated with approximately a 37% higher risk of stroke. Now, before somebody clips that sentence and we all start screaming at each other on the internet, let's actually talk

13:16 about what it means. Because 37% higher risk doesn't mean 30% of cannabis consumers are going to have a stroke. That's not what relative risk means. And an association also doesn't automatically prove that cannabis directly caused every additional strokes

13:34 observed in these studies. The researchers themselves noted substantial differences between the studies they analyzed and potential limitations in parts of the observational evidence. But they didn't stop with observational studies. They also used something called mandelian

13:54 randomization. Very simply, this uses genetic variants associated with certain behaviors or disorders to help researchers investigate whether an observed relationship may actually be causal rather than simply being explained by other lifestyle factors.

14:13 And according to the researchers, genetic liability to cannabis use disorder was associated with an increased risk of stroke overall, particularly large artery stroke. So, no, this isn't something I think we should casually dismiss, but I also

14:31 don't think the responsible headline is cannabis causes strokes. No, science is usually much less satisfying than that. It gives us evidence, then better evidence, then conflicting evidence, then another study, and eventually hopefully a clearer picture. In fact, previous

14:50 research on cannabis and cardiovascular risk hasn't always produced identical conclusions. Some studies have found increased cardiovascular risks, while others have struggled to establish significant associations once other factors are considered. And that's

15:08 precisely why studies this large matter. They add another piece to the puzzle. But here's what I find particularly interesting about the reaction to stories like this. Cannabis has spent decades trapped between two completely

15:26 ridiculous extremes. On one side, cannabis was portrayed as this uniquely dangerous substance that would apparently destroy society. On the other, you occasionally hear cannabis discussed as though because it's a plant, it must somehow be

15:45 harmless. Why do we need either extreme? Alcohol can be enjoyable and harmful. Prescription medication can save your life and have serious side effects. Caffeine has benefits and risks. We seem perfectly capable of holding two ideas in our heads at the same time with

16:04 almost everything else. So why does cannabis have to be either miracle medicine or public enemy number one? Maybe normalization actually means allowing cannabis to become boring enough that we can discuss it honestly. Benefits where the evidence supports benefits, risks where the evidence

16:23 supports risks, and uncertainty where the science simply isn't finished yet. Because if cannabis really is becoming part of mainstream medicine, consumer culture, and everyday life, then research like this becomes more important, not less. We should want to know what long-term frequent use does.

16:43 We should want to know whether dosage matters, whether the method of consumption matters, whether THC concentration matters, whether age matters, whether cardiovascular conditions change the equation. Those aren't anti-cannabis questions. They're scientific questions. And

17:03 personally, I think an industry becoming normalized should make those questions at least easier to ask, not harder. So, here's the question I'd leave you with today. Are we finally mature enough to have a cannabis conversation where good news doesn't automatically make you

17:22 pro-cannabis and bad news automatically makes you anti-cannabis? Because that's the conversation I'm much more interested in. And that's really the shift. Maybe true cannabis normalization isn't when everybody accepts cannabis. Maybe it's when we stop needing cannabis to be either

17:41 completely good or completely bad and simply start demanding the best evidence we can get.