AI slop in content: the Perfectly Imperfect newsletter as an antidote

The Perfectly Imperfect newsletter has become an antidote to AI slop: in July 2026, British Guardian columnist Emma Beddington described how a daily newsletter (pi.fyi) offers relief from content that neural networks and algorithms sand down to a faceless middle.

What is Perfectly Imperfect?

The newsletter's concept is deliberately simple: one public figure per day, and a list of what they like right now. No editorial selection in favor of what is "useful" or "viral". No sponsored inserts, no "best for your life" narrative.

The newsletter's audience is mostly American artists and musicians, not necessarily known outside their niche. But occasionally issues feature world-famous names:

• Francis Ford Coppola: Hawaiian shirts and halva
• Kylie Minogue: Japanese paper washi tape and fresh wasabi
• Lena Dunham: training pigs (literally, no metaphor)

Other documented tastes include a cocktail of Aperol, milk, coffee cream and olives; a habit of cracking knuckles against the jaw; a book on sports psychology; and the practice of calling strangers with no goal formulated in advance.

Beddington, a journalist in her fifties with a British frame of reference, admits candidly that she does not know most of the names in the newsletter. But that does not stop her reading: the preferences are so specific and odd that they give the sense of a real person, not a character from a content machine.

Why does the author set this against AI content?

Beddington calls the modern internet a "depressing cesspit", a place where every site visit ends in a "cortisol shot straight to the eyes" or AI junk interspersed with protein ads.

"The more I learn about celebrities' strange enthusiasms, the more cheered I am.

So much for AI supposedly flooding us with faceless taste slop," the author writes.

The term "tasteslop" from the column's subheading is the author's own coinage, derived from "slop" (AI junk) and applied to taste: algorithmic recommendations are optimized toward a statistical middle, and all content starts to "sound the same". Perfectly Imperfect works in exactly the opposite way: the value of the content here is directly proportional to its strangeness and the impossibility of scaling it.

An important caveat: the column is not a technology critique of AI. It is a personal essay by a journalist about what she enjoys reading. Still, it voices an observation: reader interest in plainly human, imperfect content is growing precisely against the backdrop of increasing machine smoothness.

What this means

Beddington's essay records a small but telling cultural counter-movement. As the volume of AI-generated content grows, part of the audience starts to read "too polished" as a sign of inauthenticity, a signal that no living person stands behind the text or recommendation. Formats with obvious human rough edges, such as niche newsletters, personal quirks, awkward preferences and confessions of ridiculous hobbies, are becoming a lasting bet against the algorithm.

For teams building AI products and content strategies, this is a signal: automated flawlessness is not always seen as a competitive advantage. Sometimes the reader wants precisely the crooked line that a machine would never draw.

What is AI slop and why is it a problem?

It is content that algorithms have sanded down to a faceless middle. The Perfectly Imperfect newsletter pushes back by showing the real tastes of public figures without editorial selection or sponsored inserts.

What is AI slop in content?

It is content that neural networks and algorithms sand down to a faceless middle: the monotonous, predictable output of automated processing, lacking originality and human taste.

What is the Perfectly Imperfect newsletter?

A daily newsletter in which one public figure per day shares a list of what they like right now. It has no editorial selection in favor of what is "useful" or "viral", and no sponsored inserts or narratives.