Five Things: August 6, 2026

NOTE: In a nod to a Prince reference below, I was stressed out when I wrote this, forgive me if it goes astray. August has been a crazy month (IYKYK) and it’s just getting started, so let’s just jump right into these latest things!

---

_ONE

Witch Hat Atelier: The Work of Art | Shaenon Garrity

If you tell an American publisher that this is what the average man wants to read, they will punch you and then cry. But it’s all worth it, because Witch Hat Atelier is shaping into one of the great fantasy comics.

I started reading manga more intentionally a few years ago and Witch Hat Atelier quickly became a favorite. Thirteen volumes in (I haven’t read #14 yet), it’s now one of my overall favorites series ever, and Garrity’s essay — which I’d held off reading until I’d caught up on the series, so beware spoilers — does a great job not only of explaining why it’s so good, but also helpfully positioning it in the broader world of manga.

Long-running manga series can be daunting, especially when they’re still ongoing, but after the first three volumes, I didn’t have any concerns; my only disappointment is that I caught up and have to pace myself waiting for new volumes to be released!

At first glance, it may seem like a light series meant for young girls, but Kamome Shirahama has built a fascinating setting, filled it with a cast of engaging characters who rise above their archetypes, and steadily ramped up the thematic complexity with each satisfying volume — almost all of which deliver a tantalizing cliffhanger. While the storytelling and characters are great, Shirahama’s consistently beautiful artwork and creative layouts takes everything to another level; reading each volume is a multi-sensory experience that’s unique to the comics medium.

I rarely buy alternative editions of books I already own, but I absolutely have my eyes on the new Grimoire Editions!

__TWO

Thinking Partners | Daniel Ritzenthaler

I want to brainstorm ideas with other design and product leaders. I want to intellectually spar through my plans and approaches with them. I want to rehearse my reasoning for presentations with them. It’s not that “AI” tools can’t help me do these things or that I don’t know how to use them. It’s that doing it together is the whole point. Not only for me, but for everyone involved.

I’m equally fascinated and disturbed by people who use AI tools for “ideation”. There’s a part of me that understands the appeal and has even occasionally found it tempting, but a much larger part of me sees it as one of the most insidious use cases for the “democratizing creativity” nonsense so many AI advocates love to claim for their favorite tools.

Thinking, researching, brainstorming, and workshopping are all critical parts of a democratic creative process — in the service of art AND commerce — and sparring with a computer program designed to simultaneously flatter and flatten you shouldn’t ne the least bit appealing once you give it a moment of critical consideration.

I’ve been fortunate (so far) to not have any AI tools forced upon me in a work setting (beyond them being shoved into almost everything I use), but I have worked with too many people now who have come to rely on them to varying degrees, and the experiences have varied from awkward to embarrassing. (Hello, Hank Green.) Ritzenhaler makes a particularly good point about the responsibility leaders have “to build an environment where people can learn from each other without feeling insecure,” and that starts with not making these tools a core part of their own “creative” identities.

___THREE

AI just cloned your business model | Zach O’Brien

The publishers who lose to AI will be the ones who were already publishing like machines: anonymous, generic, interchangeable. The ones who win will lean hard into the thing they’ve always had and the robots never will. They know their people.

The hand-wringing about the decline in organic search traffic and AI being the final nail in the coffin of digital media masks a fundamental truth that goes all the way back to the SEO-driven heydays of FluffPo and Buzzfeed: most online content has always been bad, written for robots, overvalued by investors and advertisers, and run by media executives who are literally serial failures by design.

It’s been interesting watching the recent AI-related scandals in book publishing as some authors have been accused of using AI to (help) write their books, but so many of the “tells” are hallmarks of “good” and “bad” writing that have been acceptable for many years, or the writing is “too good” to believe a specific writer wrote it themselves.

Meanwhile, we’re constantly being told “AI is inevitable” and everyone should strop resisting and let it happen, and some publishers are signing licensing deals with some AI companies, and some authors who openly admit to using AI tools to help them write are rewarded with attention and book deals, but also most AI tools are built on stolen content and/or exploited labor, and there are multiple active lawsuits against some AI companies… O_o

That’s why I’m just focused on being pro-craft, sitting this one out until the dust settles.

____FOUR

LemonLime wanted candidates with skin in the game. They meant it literally. | Jeremy Garland

At no point between the “fun idea” and “live event” did anyone ask the one question that mattered: how does this land on someone weighing a permanent decision against a shot at a guaranteed interview, especially in a job market like this one? That requires an informed read on power and culture, and it has to happen before an idea ships, not after it goes viral.

It’s a rough time to be looking for a job in most industries, and hustle culture has turned looking for any opportunity to stand out into a series of unfortunate events you’d understandably assume were from The Onion. In a twist on my usual format, I’m spotlighting Garland’s take on the story instead of my own, because he nails it so much better than I would have, so definitely click through and read it.

PS: I would have normally linked to William Hicks’ more traditional coverage for The San Francisco Standard, in which he describes the offending company as “One of San Francisco’s many difficult-to-explain AI startups…”

_____FIVE

Seven Takeaways From Pope Leo XIV’s AI Manifesto | Andrew Albanese

To speak of justice requires questioning the global distribution of power that decides who in fact can train these models and who is merely subjected to them. Likewise, it means acknowledging that social justice is not only a goal to be safeguarded after technologies are deployed, but a condition that must shape their very design from the outset.

I’m not Catholic and I don’t know much about the Pope beyond one of my favorite Prince songs, so I wasn’t planning to read or link to his AI manifesto that made the rounds a few weeks ago. Thankfully, Albanese took one for the team.

I’d add an eighth takeaway to his insightful take: the Pope is basically an AI moderate, but he’s diplomatically expressing the same concerns informed AI skeptics have been spotlighting for years, but have been generally dismissed as Luddites by people who don’t actually know anything about the actual Luddites.

______BONUS