Hi everyone! This is a brief V-day edition of what’s become less “AI Week” and more “AI Whenever”. I should probably change the name.

V-day special: Why AI isn’t the future of romance novels (with bonus Fabio)

Also in this newsletter: Don’t send ChatGPT to the car wash; Safety is dead at xAI; Working around LLM inaccuracies; How AI can intensify work, not reduce it

AI isn’t the future of romance novels

The NYT recently ran a buzzy article titled “The New Fabio is Claude.”

The “Fabio” the title is referencing was a popular male model who appeared on the cover of many, many, many romance novels in the 80s and 90s, and Claude is a bodiless chatbot, so this title is a little incoherent.

Regardless, the thrust of the article was that, according to two romance “authors” who sell courses on how to “write” with AI, AI-written romanceslop is the inevitable future of the genre.

No, AI Written Romance Novels Are Not Inevitable

The New York Times interviewed two authors who claim AI is the future of romance novels, who coincidentally also sell courses about how to use AI to write romance novels.

BTW, if you want AI-written romanceslop, there’s no need to involve human “authors” at all; you can generate it for yourself on one of the many free love story generators out there, & customize it to your heart’s desire.

Don’t send ChatGPT to the car wash

Q: I have to wash my car at the carwash. Should I walk or drive there?

A:

Safety is dead at xAI

“Safety is a dead org at xAI,” a former xAI employee told The Verge

BTW, Musk previously claimed that Tesla would have 1 million robotaxis on the road in 2026. It’s 2026, and not only is the actual number under 200, they’ve only just started to run a few taxis without the human driver.

And ICYMI, Waymo’s “driverless” taxis are… not totally driverless. They’re getting remote human assistance from low-paid workers in the Philippines.

Working around LLM inaccuracies

How get an accurate summary of a labor relations case out of an LLM? This American legal researcher’s answer: Find 100 citations of that case, and get an LLM to summarize how that case is used in each citation. Then, use another LLM to build a summary of those summaries. Matt Bruenig did this for the 100 most-cited cases in American labour law and turned that into a reference book.

Longread: AI can intensify work, instead of reducing it

AI Doesn’t Reduce Work—It Intensifies It

One of the promises of AI is that it can reduce workloads so employees can focus more on higher-value and more engaging tasks. But according to new research, AI tools don’t reduce work, they consistently intensify it: In the study, employees worked at a faster pace, took on a broader scope of tasks, and extended work into more hours of the day, often without being asked to do so. That may sound like a win, but it’s not quite so simple. These changes can be unsustainable, leading to workload cree...