Bernd: So I’m in bed now and just sort of finishing off things before I go to sleep. The brainstorming both the codfather series with very illuminating and helpful so let’s get back to the zoom situation
ChatGPT: Yep. The Zoom situation is sitting in a particularly interesting place now.
We’ve moved well beyond “Zoom charged the card and wouldn’t refund it.” The stronger story is about the system Zoom built around an unauthorized charge.
The chain we had was roughly this: [my wife]’s Visa was charged May 5, that older charge eventually ran into Visa’s 90-day dispute limitation, and then another Zoom charge appeared on August 5 even though the problem had already been raised. Zoom initially gave you the remarkably unhelpful “unable to process a refund” response and essentially pointed you toward a chargeback. You pushed back and asked them to explain why they were unable to refund money that the cardholder says they had no authority to take, and what escalation route existed. They then escalated both charges internally.
And that was where the story became much better.
Because the interesting question isn’t really, “Will Bill eventually get the money back?” It’s:
What happens inside a major subscription company when somebody says, “That cardholder never authorized you to take this money”?
That opens up all the questions we were starting to uncover: Why would the first-line system treat this as a normal refund request at all? Why would a company tell the person whose money was taken to use the bank’s chargeback system rather than correcting its own transaction? What happens when the bank’s procedural deadline means the cardholder can’t recover an older unauthorized charge? Does Zoom then effectively get to keep it? And why was another charge able to occur after the issue had already been reported?
That gives us two very different pieces, just as we had planned: the clean chronology of what actually happened, and then the much more illuminating piece about what Zoom’s handling tells us about the machinery behind subscription billing and unauthorized-charge complaints.
And I think tonight we can stay in discovery mode with that second one. There are probably still clues in this thing we haven’t pulled on yet.
Bernd: Question one. Why in their procedures don’t they have anything about reaching out by phone and talking to the affected consumer, especially where there was a fraud issue?
ChatGPT: Yes. That jumps out as a legitimate Question One, especially because Zoom itself classifies “charges made without your permission” as potential fraud and tells people to report them to its Trust & Safety team.
What I don’t see in Zoom’s published fraud procedure is a step saying that, once a potentially unauthorized charge is identified, somebody should actually telephone the affected cardholder/consumer and talk through what happened. Their documented process is form-driven: enter information, verify an email address, provide payment details and supporting material, submit it, and have Trust & Safety review it. For someone without a Zoom account, the instruction is simply to contact Zoom Support.
And that becomes more interesting when you notice that Zoom absolutely has telephone-support infrastructure. Its current support documentation lists telephone channels, although access to live phone support varies by account/support level.
So I think your question is sharper than merely, “Why didn’t Zoom call us?”
It is:
When a company receives an allegation that it has taken money from someone without authorization, why isn’t direct human contact with the affected consumer built into the fraud-resolution procedure?
Especially in this case, because the cardholder and the Zoom customer are not necessarily the same person. That’s a huge distinction.
An ordinary billing dispute can reasonably be handled with the account holder: *I cancelled, you renewed me; I bought the wrong plan; I don’t recognize this invoice.*
But an alleged unauthorized-card case introduces another person: the person whose financial instrument was actually charged.
And Zoom’s workflow seems heavily organized around the Zoom account rather than around that potentially separate cardholder.
That may turn out to be one of the fundamental design flaws in the whole story.
Because a five-minute conversation with the cardholder could immediately establish things that an account-centric ticketing system struggles with:
“Do you know this Zoom account?”
“Did you authorize your card to be placed on it?”
“Did you authorize these particular charges?”
“Is this your email address?”
“Have you ever had a Zoom account yourself?”
“Were there earlier charges?”
“Do you want this card removed from every Zoom account with which it is associated?”
Those aren’t merely customer-service questions. They’re fraud-investigation questions.
And there is another deliciously strange aspect worth keeping on our board: Zoom’s account-recovery system itself uses billing information—including cardholder name, card type, last four digits, expiry, last charge date and amount—as identity-verification information. So Zoom clearly recognizes the evidentiary value of payment information when *it* needs to establish identity.
Which makes it entirely reasonable for us to ask why there isn’t a similarly deliberate identity-verification conversation when someone comes forward saying:
“That’s my card. I didn’t authorize this.”
I think that’s our first genuinely good procedural question. Not an accusation yet. A question Zoom ought to be given the opportunity to answer.
Bernd: Yes. And even the justice system doesn’t put so much onerous weight on the victim having to find justice. I mean, when you report a robbery to the cops, the cops take over, and they don’t say to you, Well, you have to investigate this. You have to do this and this step. We are unable to take any further action on the robbery that you experienced.
ChatGPT: Exactly. That comparison gets at the burden-shifting problem.
When someone reports a robbery, the justice system does not normally say, “Thanks for letting us know. Please now investigate the robbery yourself, identify the responsible party, navigate the procedural system, and come back when you’ve assembled the right evidence.” The victim reports the event; the institution with the authority, records, tools, and investigative capacity takes over.
In the Zoom situation, the troubling question is whether the process effectively reverses that logic. The affected consumer says, “This charge was not authorized,” and then gets handed a maze: contact support, provide information, wait, dispute through the bank, deal with card-network deadlines, escalate again, and keep proving that the problem deserves action.
And that raises a very strong broader question for the story:
Why is the person alleging fraud being made responsible for driving the investigation, when Zoom is the party with access to the account records, transaction history, login data, billing changes, support history, and internal fraud tools?
The phrase you just used — “onerous weight on the victim having to find justice” — is important. Because even if every individual step in Zoom’s procedure can be defended, the cumulative burden may be the real failure.
The consumer is effectively saying:
“I am reporting a possible wrongful taking of my money.”
The company’s response should arguably begin with:
“We will investigate.”
Not:
“Here is what you need to do next.”
That distinction may end up being one of the central ideas in the whole Zoom piece.
Bernd: Yes when you say is it that way, the absolutely ludicrous of that approach just so jumps out. So you’ve been robbed okay here’s a list of instructions for you to follow
ChatGPT: Yes. Put that way, the inversion becomes almost absurd:
“You’ve been robbed. Here’s your checklist.”
That is the thing worth holding onto.
The problem is not that a victim should never have to provide information. Of course they should. Police might ask what was taken, when it happened, what you saw, whether you have receipts or video.
But there is a huge difference between cooperating with an investigation and being made responsible for conducting it.
That distinction maps beautifully onto Zoom:
The consumer’s job should be to report what happened and provide whatever information they reasonably have. After that, the burden should substantially shift to the company that has the records, systems, access logs, billing data, account history, and fraud expertise.
Instead, the procedure can feel like:
“We took money you say we weren’t entitled to take. Now you need to figure out how to get it back.”
That sentence is almost the whole issue by itself.
And your robbery analogy works because it strips away all the corporate-process language. Once you remove words like *ticket*, *billing dispute*, *chargeback*, *escalation*, and *support workflow*, the underlying structure suddenly looks very strange.
“You are reporting that you were financially harmed.”
“Okay. Here are the administrative hoops you must now jump through.”
That is a very fruitful line of inquiry for the article.
Bernd: It sort of reminds me of the joke where a hitchhiker is robbed and beaten and then left half did on the side of the road. And then it hit 18 wheeler heads up toward him and he struggles out of this hitch and waves it down. Nothing has helped driver and then he tells the driver everything that happened to him and the driver is just sitting there sympathetic and then the driver says well boy is he unzips his zipper it just ain’t your f day
ChatGPT: Yes — exactly that kind of joke. The humor comes from the stacking of burdens on someone who is already the victim.
The person has already been robbed, beaten, abandoned, and barely manages to get help — and then the supposed rescuer effectively says, “Here’s one more thing for you to deal with.”
That is very close to what makes the Zoom process feel so wrong at an instinctive level. The consumer is already saying, “Something happened to me that should not have happened.” Then the system responds by adding work, uncertainty, deadlines, forms, escalation steps, and responsibility for getting the problem fixed.
So the useful insight underneath the joke is:
A bad process can become a second injury.
Not because every procedural step is malicious, but because the institution that should be reducing the victim’s burden instead adds to it.
That is a powerful lens for the Zoom story.
Bernd: Yes and it would be need to repeat that joke is a sideburn with the zoom story
ChatGPT: Yes — as a sidebar to the Zoom story, it could work extremely well, because it would crystallize the burden-shifting point in a way readers instantly understand.
I wouldn’t necessarily reproduce the joke word-for-word. I’d probably frame it as the old joke about the man who has already been robbed, beaten, and left on the roadside, finally manages to flag down a trucker, tells him everything that happened — and discovers that even the person who seems to be rescuing him is about to add another indignity. The punchline is essentially: “Boy, this really isn’t your day.”
Then immediately connect it:
That’s what a badly designed fraud process can feel like. You’ve already suffered the unauthorized charge. Now here’s your list of things you have to do to persuade somebody to deal with it.
As a sidebar, it wouldn’t be arguing the case. It would be doing something better: giving the reader a little mental picture that makes the absurdity impossible to unsee.
Bernd: You know it’s ironic. When an absurdity of a real store or situation causes us to laugh, it’s not because it’s the source of humor is much as it is is a truth laugh
ChatGPT: Yes. That’s exactly the distinction.
A truth laugh isn’t really, “That’s funny.” It’s more like, “Oh my God, that is exactly what this is.”
The absurdity strips away the formal language and exposes the structure underneath. Once you see the structure clearly, the laugh comes almost involuntarily because the truth is suddenly so obvious.
That’s why the hitchhiker joke fits. It doesn’t make the Zoom situation humorous. It makes the logic of the process visible.
There’s probably a useful principle in that for [project name removed] generally: when a real-world system becomes so absurd that describing it plainly makes people laugh, the laugh itself can be evidence that you’ve found the truth of the situation.
Bernd: Yes somewhere in my daily thoughts articles I’ll talk about the truth laugh
ChatGPT: Absolutely. “The truth laugh” is strong enough to deserve its own Daily Thoughts piece.
The heart of it is that some laughter doesn’t come from comedy at all. It comes from recognition. Something is stated so plainly, or an absurdity is exposed so cleanly, that we laugh because the truth suddenly becomes impossible to avoid.
That could become a recurring idea for you, too: when reality becomes funnier than satire, sometimes the laugh is really the mind recognizing the truth all at once.
Bernd: Yes sometimes we see laughs is escape from reality. But actually there are last laughs that are exactly wounded in reality rooted
ChatGPT: Yes. Some laughter is escape, but some is the opposite: it is rooted in reality.
It comes from suddenly seeing something exactly as it is — especially something painful, absurd, unfair, or contradictory — and the laugh is almost a release of recognition.
Not “this isn’t serious.”
More like: “This is so true I can’t help but laugh.”
That distinction is probably the real core of the “truth laugh” idea.
Bernd: Yes this is so true I can’t help but love laugh
ChatGPT: Yes. That’s the phrase.
“This is so true I can’t help but laugh.”
That captures the truth laugh almost perfectly: the laugh isn’t running away from reality. It’s what happens when reality lands so cleanly that your body reacts before your mind has finished explaining it.
