Skip to content

Instantly share code, notes, and snippets.

@stvhay
Created March 22, 2026 19:55
Show Gist options
  • Select an option

  • Save stvhay/ab802be38cc3ce788fb3471800921f27 to your computer and use it in GitHub Desktop.

Select an option

Save stvhay/ab802be38cc3ce788fb3471800921f27 to your computer and use it in GitHub Desktop.
The Gist — Story Pitch for March 22, 2026 (Sunday)

The Gist — Story Pitch for March 22, 2026 (Sunday)

The Gist — Story Pitch for March 22, 2026 (Sunday)

Prepared for Mike Pesca by the producer team. Stories ranked by Pesca-fit: contrarian angles, "wait what?" factor, substance beneath the surface. Editorial scores shown as W/A/H/S (Wait-What / Angle / Humor / Substance, 0–10 each) with mechanical rank in brackets.


1. "It's Gambling" — Prediction Markets Get Their Day in Court

Editorial: 8.5 · Mechanical rank: #13

The pitch: A Nevada judge issued a temporary restraining order Friday blocking Kalshi — the Trump-administration-backed "prediction market" — from operating in the state. The ruling called it what it is: gambling. On Kalshi's "sports" tab you can "buy shares" in NCAA tournament outcomes and collect a profit if your team wins. Defector's piece dissects the semantic gymnastics head-on. Meanwhile AEI flags broader "deadly implications" of prediction markets, and the co-founder is a 29-year-old former ballerina who just became the youngest woman billionaire. The language angle is pure Pesca: when does rebranding an activity change what it actually is?

Pesca angle: Washington loves prediction markets right now — they're seen as epistemically superior truth-seeking instruments. But a Nevada court just said "this is illegal gambling." The gap between what something is called and what it does is where this segment lives.

Sources:


2. Federal Cyber Experts Called Microsoft's Cloud "A Pile of Shit." They Approved It Anyway.

Editorial: 8.0 · Mechanical rank: #12

The pitch: ProPublica investigation into FedRAMP — the federal cloud security certification program. Government cybersecurity experts assessed Microsoft's offering, concluded it was deeply insecure, one reviewer literally called it "a pile of shit," and then certified it anyway. The story is about how institutional incentives override expert judgment: reviewers knew it was bad, documented that it was bad, and approved it because the system doesn't actually empower them to say no.

Pesca angle: The system working exactly as designed, badly. Everyone in the chain did their job — reviewed, flagged, documented — and the outcome was identical to if they hadn't. It's a story about the theater of oversight.

Sources:


3. Robert Mueller Dies — and Trump Says "I'm Glad He's Dead"

Editorial: 7.5 avg (three clusters combined) · Mechanical ranks: #18, #19, #36

The pitch: Robert Mueller died Saturday at 81. Trump posted on Truth Social: "Robert Mueller just died. Good, I'm glad he's dead." The Atlantic's response focuses on the biography gap: a man who avoided Vietnam with "bone spurs" dancing on the grave of a Marine rifle platoon commander who earned a Bronze Star and Purple Heart in Vietnam. This isn't partisan rage — it's a character-study story about two men who defined an era and what they each understood public service to mean.

Pesca angle: You don't need to editorialize — the biographical facts do the work. Mueller: Princeton, Marines, 12 years as FBI director under two presidents, special counsel. Trump: bone spurs. The contrast is the segment.

Sources:


4. Musk Found Liable for Twitter Investor Losses — But Not Fraud

Editorial: 7.5 · Mechanical rank: #11

The pitch: A San Francisco jury found Musk's tweets drove down Twitter's stock price and cost investors money — but absolved him of "engaging in a scheme to defraud." So the tweets were misleading, but there was no scheme. Nine outlets are covering it, making this the most widely sourced story in today's feed. The irony: the owner of the platform where the tweets happened, found liable for the tweets themselves. Can a tweet be a legally actionable statement?

Pesca angle: The jury's split decision invites unpacking. "Your tweets were misleading but you weren't scheming" is a legal distinction that reveals something interesting about how we classify intentional behavior. It also opens the question of what it means when the world's most powerful person treats his own platform like a stream of consciousness.

Sources:


5. SNL UK Launches — Its First Cold Open Is About Not Going to War

Editorial: 7.5 · Mechanical rank: #4

The pitch: "Saturday Night Live U.K." premiered last night, Tina Fey hosting. Cold open: Keir Starmer sending Trump a break-up voice note explaining he can't join the Iran war but "we can still be chums." Nicola Coughlan, Michael Cera, and Graham Norton crashed the monologue to ask why an American was hosting; Fey's answer: "None of you f---ers would do it." Meanwhile Timothy Garton Ash in The Guardian is writing about real US-UK contempt, and Farage is trying to distance himself from Trump. The show and the reality have merged.

Pesca angle: Can you franchise a comedy institution? SNL's formula is so American — the cold open, Weekend Update, the host monologue — that exporting it is a cultural experiment. The fact that the first political sketch is about the UK saying "no" to America tells you everything about where the relationship stands right now.

Sources:


6. $800 a Day to Bully AI Chatbots: The New Service Economy

Editorial: 7.5 · Mechanical rank: #22

The pitch: An AI memory startup called Memvid is paying $800 for a single eight-hour shift — $100/hour — for someone to spend the day telling AI chatbots what to do on camera. No special qualifications required beyond willingness to be bossy. Tyler Cowen at Marginal Revolution flagged this under his ongoing "those new service sector jobs?" series on weird labor market developments. What do "jobs of the future" actually look like? Not coding — just yelling at robots for a living.

Pesca angle: The absurdity is the hook, but there's a real question underneath: is the AI economy creating new kinds of work, or just new kinds of weird? The "bully the robot" job posting is both a genuine economic data point and a perfect encapsulation of how confused our intuitions about AI jobs still are.

Sources:


7. Publisher Pulls Horror Novel "Shy Girl" Over Suspected AI Authorship — Writer Denies It

Editorial: 7.5 · Mechanical rank: #53

The pitch: Hachette yanked horror novel Shy Girl from U.K. shelves and canceled the U.S. release after "multiple allegations of AI use." Ars Technica calls it "one of the first controversies of its kind." The author denies it. The detection problem is where this gets interesting: how did they catch it? What does "suspected" mean — is there a reliable test? And what about authors who just write in a stilted, mechanical way naturally? Hachette cited commitment to "original creative expression," but they can't define it.

Pesca angle: Publishing is where the "is this real?" question gets existentially weird. The question isn't whether AI writing is bad — it's whether "AI-written" is even a meaningful category when the human-written version might be indistinguishable. If a human writes something that reads like AI output, is that the human's problem or the detector's?

Sources:


8. Trump FCC Lets Nexstar Blow Past the 39% TV Ownership Cap

Editorial: 6.5 · Mechanical rank: #15

The pitch: FCC Chairman Brendan Carr approved Nexstar's $3.5 billion acquisition of Tegna, explicitly letting the combined company exceed the national 39% station ownership limit — a cap that has existed for decades to prevent media concentration. Eight state AGs immediately filed for a TRO. The cap isn't being repealed — it's being waived for one favored broadcaster. That's a more interesting story than straight deregulation.

Pesca angle: "Rules that exist but don't apply to you" is a Pesca sweet spot. The 39% cap still exists for everyone else. The rule is now functioning as something other than a rule — it's a rhetorical fig leaf.

Sources:


9. DHS Shutdown — TSA Workers Are Quitting, and Trump Threatens ICE at Airports

Editorial: 6.5 avg (three clusters combined) · Mechanical ranks: #6, #16, #45

The pitch: Five weeks into the DHS funding standoff, TSA officers are skipping shifts or quitting because they're not paid. Airport lines are stretching to multi-hour waits during spring break. The Senate's Saturday funding gambit failed 41–49 on party lines. Now Trump has threatened to deploy ICE agents as airport security if the standoff continues. The escalation arc is the story: a political impasse has now produced a real-world threat to use immigration enforcement as airport security.

Pesca angle: The mechanics of what actually happens when a federal agency runs out of money for five weeks — not the politics, the operational reality. And the ICE-at-airports threat is the kicker: a problem created by a funding dispute is being offered a solution that would create entirely different problems.

Sources:


10. DOJ Sues Harvard for Billions Over Antisemitism — But What's the Real Play?

Editorial: 6.5 · Mechanical rank: #25

The pitch: The Trump administration sued Harvard Friday, alleging the school was "deliberately indifferent" to harassment of Jewish and Israeli students after October 7th. The DOJ is seeking billions. Axios frames it as the latest use of antisemitism investigations as leverage against universities. NYT is tracking how Columbia responded to its $400 million ultimatum a year ago — promises mostly kept, not all. The antisemitism concerns are real AND the government may have other motives. Both things are simultaneously true.

Pesca angle: When the government weaponizes a legitimate complaint, does it help or hurt the original cause? Jewish students who genuinely experienced harassment now have their grievance entangled with a power play. That's the tension — not "is Harvard antisemitic?" but "does this lawsuit actually protect Jewish students?"

Sources:


Filed March 22, 2026. 877 articles fetched, 1,265 clusters formed. Editorial scoring: W = "Wait, what?", A = Angle/contrarian potential, H = Humor/absurdity, S = Substance (0–10 each). 60 mechanistically ranked clusters evaluated; ~25 immediate disqualifies (pure sports results, breaking news without angles, misclustered content, miserabilism). Top 10 selected from remaining scored candidates.

@stvhay

stvhay commented Mar 22, 2026

Copy link
Copy Markdown
Author

post-hoc 60

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment