In This Issue

Every election cycle, a certain type of argument makes the rounds on what exactly Democrats need to do to win in November. The advice typically includes some variation of: move to the right. Ignore your base, and aim for the center. Triangulate until you have no discernable policy positions or personality of your own. 

Much of this advice comes from centrist or Republican pundits — what the podcast Citations Needed called “the Inexplicable Republican Best Friend,” always on hand to offer guidance to their political opponents. (As co-host Adam Johnson put it, it’s like someone saying, “I’m an ice cream man and I think the solution is to buy more ice cream.”) This week, we got a special new case in The Atlantic courtesy of Nathaniel Frum (son of Atlantic editor and former George W. Bush speechwriter David), who argued that Dems just need to learn how to talk about sports. Below, we run through some classics of the genre. 

Be More Relatable to the Common Man

By learning how to talk about sports

Per Frum, Texas Democratic Senate nominee James Talarico could “prove that he’s a regular guy” and get positive attention by dunking on Arch Manning, scion of the uber popular Manning football family.  

By going on more bro-y podcasts

There was clearly one simple diagnosis for Kamala Harris’ crushing defeat to Trump in 2024: Democrats had to find a “Joe Rogan of the left” and win back young (white) men. 

Stop Supporting Things You Don’t Support

Like defunding the police

Democrats are often accused of being weak on crime and anti-police — stances that purportedly cost them votes in the 2022 and 2024 elections. But a vanishingly small number of elected Democrats ever argued for defunding the police (it’s worth noting that those who did also made the case for reallocating that money towards public safety initiatives). Famously, Joe Biden earned loud applause at the 2022 State of the Union with his call to “fund the police.” 

Like being so dang woke 

Self-professed liberal Bill Maher is among the many voices often calling for Democrats to ditch their “woke” obsession with allowing trans women to play women’s sports and “put[ting] race at the front of everything.” In reality, Democrats rarely campaign on promoting trans rights, and prominent figures in the party from Rep. Seth Moulton (D-MA) to California Gov. Gavin Newsom have adapted GOP talking points when speaking out against trans women and girls being allowed to play women’s sports. 

Promote Republican Policies and Politicians

Like anti-abortion candidates

New York Times columnist Ezra Klein thinks being a big tent party means running and supporting anti-abortion candidates, in a post-Roe era when reproductive healthcare is all but inaccessible in large swaths of the country. Klein made this case last fall, even after ballot measures protecting or expanding access to the procedure passed and drove massive voter turnout in states across the country. As Jessica Valenti put it, “To single out abortion, of all things, as the place for compromise is to ignore the political reality of the last three years.”

[Essay]

by Derick Dirmaier

This week, I finally installed the AI detection software Pangram in my browser. For months, I’d been intrigued by social media users who’d post about analyses they’d done on suspected AI text. Then there were the AI scandals that made headline news: Hatchett’s cancelled publication of the novel Shy Girl after Pangram showed it was likely to have been written with AI, the pope’s X account flagged by Pangram for a suspected AI-generated tweet, a Modern Love column in the New York Times determined to be AI by a researcher using Pangram. At first the narrative appeared relatively positive. With much hand-wringing over the future viability of human-written text in a world of AI, here was finally a tool humans could use to fight back. Previous AI detection software had been spotty enough to apparently determine the Declaration of Independence was AI. Pangram, meanwhile, boasts an accuracy rate of 99.98%. In the hands of responsible administrators, it could potentially offer an effective deterrent to AI writing — not all that different from the anti-plagiarism software of the early 2000s. But in a digital world supposedly awash in AI-generated text, what would happen if everyone had access to this tool? What if we all started using Pangram to do our own analyses of what is AI and what is human? And how does this thing even work?

Pangram is itself an AI, part of a long tradition in Silicon Valley of offering up more technology as a solution to problems their technology created. The simple explanation of how it works is that it is trained on datasets of both human and LLM-produced text to determine the probability of each individual input (e.g. word or punctuation mark) belonging to a human or AI. Based on the probability of each input appearing in particular order over the course of the entire document, Pangram provides a likelihood that the author is human, AI, or somewhere in between. For example, if a human might use an em dash 50% of the time, an LLM might use it 90% of the time. The more inputs you have to compare, the more accurate the detection. And Pangram is, by most accounts, accurate. The company claims a false positive rate of only 1/10,000 and my own efforts to trick it have so far been unsuccessful. It correctly identified all 10 writing samples in the New York Times AI vs human writers quiz, something I could annoyingly not do. Even running AI-generated texts through “humanizers” — AI tools used to replace LLM tropes with more natural language — didn’t help avoid detection (others have apparently had more success). This all seemed promising, but I couldn’t shake the feeling that using Pangram to detect AI text wasn’t all that dissimilar to using an LLM to generate text. In both cases I was dropping text into a black-box and waiting on the results.

With Pangram installed in the browser, the experience of using the tool changes. Instead of copy and pasting text into the app, you simply highlight and right click on what you’re reading for an on-the-spot analysis. When you browse social platforms like X, Substack and Reddit, that analysis is automated so that users and posts are identified as human or AI as you scroll through the feed. This might be as dystopian as Richard Deckard scanning for replicants, but it’s also boring and not the least bit empowering. We already know that our feeds and search results are stuffed with spam and slop, whether AI-generated or not. AI detection is only really interesting when there is a human on the other end, either intending to deceive or being accused of deception. And it’s here where the narrative around Pangram starts to shift.

Last Sunday, the Atlantic ran a headline stating that America Has A Pangram Problem. Considering that the Atlantic was one of the foremost publications to buy into the benefits of Pangram, this caught my intention. The problem, according to writer Matteo Wong, is that Pangram’s accuracy has emboldened users to conduct witch hunts for AI using Pangram and cite the results as evidence. He points to a recent case where journalist Taylor Lorenz was accused of using AI after a Pangram scan, only to be later vindicated when she was able to show her edit history (ironically, Lorenz had recently declared Substack inundated with AI after she had performed a Pangram analysis of the platform). Wong also mentions efforts by Pangram users to highlight passages in Pope Leo’s AI-skeptical encyclical that may have been written with AI, accusations that the Vatican, of course, denies. 

In addition to the bad-faith actions empowered by Pangram, there are also technical and conceptual problems with AI detection. LLMs evolve quickly, always toward the aim of appearing more human. It’s unlikely that tools like Pangram can keep pace, a prospect Wong likens to “building a sandcastle at low tide.” Conceptually, AI-detection might be too far downstream of where the biases of the LLM have the most influence on the writer, like during research. Finally, there is the question of how much people even care that something was AI-generated. It might feel rude or icky to encounter AI during what you thought were personal exchanges, but when it comes to art, studies continue to show that people prefer AI to humans — a finding writer Max Read explains by suggesting that people just like bad art, which seems undeniably true.

What Pangram appears to have accomplished is at least temporary proof that AI can be effectively used to identify other AI. But the product marketed to users has a slightly different objective. If Pangram really wanted to rein in AI bot slop, it would sell to the platforms, not the users. The reason that they don’t isn’t just that most platforms have little incentive to filter out AI, it’s that what Pangram is selling isn’t AI detection but agency. The AI companies have convinced us that AI is smarter than we are and thus a convenient tool for deception. Pangram sells us a false security of being able to root out deception.

In F is For Fake, Orson Welles’s excellent documentary essay on fakery, Welles explains that as long as there are fakers, there have to be experts. He then asks: but if there weren’t any experts, would there be any fakers? I doubt we’ll ever find out.

[Good Twetes]

[Words of Wisdom]

Progressive political commentator and strategist turned Unusual Take Haver Briahna Joy Gray: “I believe myself to be COVID vaccine-injured.”

[This Effing Guy]

Congressman Andy Ogles may be an avowed Islamophobe, but don’t you dare accuse him of being homophobic. The Tennessee Republican got into some hot water this week for kicking off Pride Month with a post on X declaring that “homosexuality has no place in America.” After bipartisan criticism, Ogles deleted it, blaming a comms staffer for the message, which he called “stupid, hurtful and a complete distraction from my America First focus,” and claimed he only heard about “while working on [his farm].” The old “scapetern” excuse might have been more effective if Ogles didn’t have such a long record of making inflammatory statements on social media. In March, for example, he posted, “Muslims don’t belong in American society. Pluralism is a lie.” 

[In the Cafe]

In the lead-up and aftermath of his appearance at the “Remigration Summit 2026” in Portugal last weekend, former Customs and Border Protection commander Greg Bovino gave interviews to:

• Far-right website VoxEuropa, in which he cited Nazi Germany general Erwin Rommel as an inspiration
• Irish white nationalist influencer Keith Woods, a self-described “raging antisemite.” 

So you can imagine the kinds of people who actually organized and attended the conference. Freelance journalist Christopher Mathias broke it all down for TPM in a new piece this week, and joined us on Substack Live to talk more about Bovino’s whole deal.

[TPM Trivia]

1) Who is the CBS employee fired this week for speaking out against Bari Weiss’ takeover of the news network?

2) What jobs does Bill Pulte hold in the second Trump administration, including a new one announced just this week?

3) What is the name for the (confusing) style of primary election they use in California? 

4) Which former Trump official this week agreed to a plea deal over his retention of classified information?

Answers below

[No Words]

NEWARK, NEW JERSEY – MAY 31: A protester waves an upside-down American flag at a police blockade near the Delaney Hall detention center during a protest against the transfer of detainees and federal immigration policies on May 31, 2026 in Newark, New Jersey. Tensions remain high outside the detention facility where activists have clashed with police for days after detainees began a hunger strike over Memorial Day weekend amid allegations of inhumane living conditions. (Photo by Andres Kudacki/Getty Images)

Trivia Answers: 1) Scott Pelley of “60 Minutes” 2) Director of the Federal Housing Finance Agency and the chairman of Fannie Mae and Freddie Mac. Now also Acting Director of National Intelligence. 3) Jungle primary 4) John Bolton