How Cannabis Creators Beat Social Media Censorship — One ‘Broccoli’ at a Time

Social media platforms have become the new censors of cannabis culture. Josh Kesselman posts at 2am India time. Edible Dee rewrote four books. Riley Cannabichem puts on a metaphorical white coat. Here’s how the sharpest minds in cannabis content are surviving the algorithm — and what they’ve learned the hard way.

Here’s a quick exercise: imagine you’re at a penthouse party—pool, great guests, great vibe. A Velvet Underground record is playing in the background (or an old reggaeton playlist, it doesn’t matter), and you have to explain to someone what you’re “carrying” without triggering the algorithm, that digital bouncer running on HAL 9000 logic. The problem? At this event, there are no written rules: the bouncer will kick you out if you say “marijuana,” but might let you continue enjoying the party if you spit out “grass,” “broccoli,” or “lettuce.” A logic that’s hard to predict.

That’s how things are on the internet today—this silicon Matrix run by bots (those snitches programmed in zeros and ones) where Mark Zuckerberg, Google, and Chinese tech platforms have donned their cop caps, and social media plays at being a modern-day Inquisition. That’s why, for those of us hardened by the ups and downs of cannabis culture, posting a picture of a joint can be a risky business: one day you’re the king of engagement with thousands of likes piling up; the next, you wake up to find your account shadowbanned, and good luck complaining to who-knows-who.

But as the saying goes, “Where there’s a will, there’s a way.” And in this chess game against censorship, those in the know move stealthily like the ronin in The Lone Wolf and Cub. So, to put it bluntly: the gist comes down to semantics, to that shared code that makes us feel part of a kind of 420 club. That makes us stand on the same side. That allows us to understand each other even amidst the cynical contortions and pirouettes of language.

The pretzel problem

Josh Kesselman, owner of RAW and publisher of High Times, sees a solution: the key—always—is originality. “You’d be surprised how inventive you can be,” Josh reflects. He gives us the scoop on “pretzels” or talking about “making a delicious burrito” while the camera shows something else. However, he cautions that no tactic lasts forever: “I came up with years ago on TikTok using pretzels. And it worked at first. I was getting millions and millions of views, but then other people copied me and took one too far with it. I made it obvious that AI learned what that pretzel meant to something else. And then I couldn’t use pretzels anymore for a while. The views went down to almost zero as soon as I used a pretzel. You move on to something else.”

So Kesselman’s recommendation is to play with the gaps: “Skip words, just infer things. Skip the word. Say it in a funny way. Say anything,” he advises. “You can always just blanket it. When you’re enjoying, just… and skip the word. You can just completely skip it. That always works too.” The key is to let the viewer complete the sentence and find a knowing wink so the idea falls into place naturally. It’s pure pop ingenuity, like back in the ’70s when people used code to avoid drawing DEA attention.

For Josh, visual literalism is another weapon: “You can actually show an actual broccoli.” This technique extends to other elements: “I’ve done videos with actual green tea and I make it clear. You can’t just use green tea. It has to say on there in big letters, GREEN TEA, so everyone knows it’s green tea. Same thing with the pretzels. I’m using pretzels. I have big boxes of pretzels right in front of you, so you can see exactly what I’m using.”

When the algorithm wins — and you have to start over

Danielle Russell, also known as Edible Dee, the Happy Chef, a celebrity cook on the WWW, had to rework her entire catalog. After years of battling Meta’s censorship, she decided that if you can’t beat the algorithm, you have to dress up as Clark Kent. Now, her books talk about “infusing with happiness” or using “magic” instead of throwing around technical terms that trigger the alarm bells of digital censors.

“To bypass censorship, focus on using alternative platforms, fostering collaborations, and employing process/science language over consumer/lifestyle language where possible. As a four-time author, I have also—because of the losses I have personally suffered due to these platforms—had to release second editions of all my Updated works, updating the active ingredient verbiage.”

Danielle Russell (Edible Dee), The Happy Chef

The white coat strategy

Not everyone hides in the shadows of euphemism. There are those, like Riley Cannabichem, who prefer the white coat strategy, very much in the style of Beakman’s World. She stands in front of the camera and uses the real words—cannabis, THC, CBD—but with a protective shield of science and education. Interestingly, the algorithm seems to have an almost mystical respect for academia.

“I’m typically talking about research studies or fun facts about the plant—all backed by science—and I think the algorithm sees the scientific basis and allows it to stay, but who knows what’s really going on.”

Riley Cannabichem, researcher and content creator

So, if you talk about research studies and keep the joint off-camera, censorship might let it slide.

Fighting fire with fire — and AI

Natalia Kesselman, High Times deputy editor, editorial director of El Planteo and one of the minds responsible for shaping the media outlet’s digital strategy, proposes a kind of technological guerrilla warfare: fighting fire with fire, like in a WarGames hacker duel. She revives timeless classics like “magic broccoli” and suggests that the aesthetic of blur or out-of-focus is not just an artistic choice, but a survival necessity. But the most disruptive aspect is her use of artificial intelligence.

In her words: “Once you’ve built the foundation of your content, you have to feed it to the AI for a security review. You have to tell it strictly not to touch anything you’ve done, but ask it to create a ‘reliability index’ based on how explicit everything is and to point out which terms or images could generate conflicts or raise red flags in the algorithm. It’s a truly spectacular tool.” ChatGPT as a security consultant infiltrated behind enemy lines.

The full playbook

This isn’t paranoia — it’s a response

Finally, Josh Kesselman reveals that this level of caution isn’t paranoia, but a response to direct attacks: “I truly believe that someone pays to try to keep my posts down, to try to get reported and all these things. And so I have to be more conservative than most. You wouldn’t think so, but I have to be. It’s all these crazy things you see me doing. These are all like very well thought out.”

He even resorts to time zone tactics: “Sometimes I post at odd times because I know that if they’re paying somebody in India, for example, I’ll post at 2 a.m. India time. Because that way people are not awake. I’ve really tried everything I can to work around their attempts to keep us down. And I’ve done a great job. We still grow. And it shows.”

At the end of the day, it’s about remaining authentic in a world that wants us to be uniform, neat, docile, and weed-free. Because even though they try to silence us, cannabis culture always finds a crack for the green to peek through, even if today we call it “happiness,” “pretzels,” or “magic broccoli.” “It’s a matter of being inventive,” Josh concludes. Because they can prohibit the word, but they’ll never be able to clear the smoke.