Digital Exhaust #196
Pseudo-profound bullshit and chatbots as medical devices
Thanks for opening. A few things I found this week:
Penn Medicine pulls out
Penn Medicine said enough to U.S. News and World Report rankings. They announced this week that they would no longer be participating.
This is major news as they are one of the first health systems to play chicken with the The Man. A handful of medical schools pulled the plug on the esteemed ranking system earlier this year. As I suggested then, this system is configured such that everyone has to play. Otherwise it’s worthless. A couple of more Ivy hospitals fall and the whole thing will cave in.
Maybe the most interesting thing here is how little press the story received.
Research on profound bullshit
Remarkable. Academics are actually studying bullshit. I stumbled on this recent study from the Cambridge University Press:
Here we focus on pseudo-profound bullshit, which consists of seemingly impressive assertions that are presented as true and meaningful but are actually vacuous. We presented participants with bullshit statements consisting of buzzwords randomly organized into statements with syntactic structure but no discernible meaning.
This may be one of the most important studies ever published.
Overheard | Commodity frontier
The threshold of activities we deem “too personal to pay for.” It's a perimeter that exists even when the products we enlist are free, and the arrival of new technologies calls for its constant renegotiation.
I found this in a wonderful essay addressing the question, ‘should I use and AI to write my wedding toast?’ The author, Meghan O’Gieblan, unpacks the issue in the most nuanced way. We will all participate in this kind of negotiation with AI.
+ O’Gieblan is the author of God Human Animal Machine, one of my fav reads of 2022.
AI chatbots are medical devices
Excellent Nature Medicine review of the complex landscape of healthcare LLM regulation. In a world moving fast and loose with OpenAI in healthcare, this offers a sobering breakdown of why we need to pump the breaks and view these products as medical devices. It’s well-written and approachable for those without deep knowledge of AI.
An LLM can provide convincing advice, but cannot tell the patient where the advice came from, why it is giving the advice or how carefully ethical balances have been considered. To earn their place in the medical armamentarium, chatbots must be designed for better accuracy, with safety and clinical efficacy demonstrated and approved by regulators
Thanks to Hugh Harvey (author) for lending Digital Exhaust the link behind the paywall. If you want to understand how AI will integrate in medicine without the hype, give him your attention.
On the go buddhism
Or how Silicon Valley bastardized mindfulness.
Behind the paywall at The Atlantic I found an excellent essay on the experience of meditating with Calm — Calm, a popular meditation app, turns the slightly unaccountable act of meditation into something I can accomplish, and cross off the list. The piece subtly captures the irony of using your smart phone for mindfulness practice.
This reminded me of an excellent book I read, Work Pray Code — When Work Becomes Religion in Silicon Valley, which captures the whole meditation cottage industry nicely:
These apps train users to think of meditation as a consumable tool with discrete functions, like a hammer, a screwdriver, or a wrench…. Meditation apps allow users to monitor their "progress" while meditating, turning meditation — a practice closely associated with the ideal of non-attainment — into a goal-oriented activity.
On the go Buddhism seeks to capture and tame the sacred but winds up killing it.
+ Take this to the bank: The problems created by technology can’t be fixed with more technology.
Medical schools eye workarounds to SCOTUS ruling
The Supreme Court ruled that higher education can no longer take race into consideration as a basis for granting admission. This overturned a long-standing precedent that benefited Black and Latino students in higher ed. Medical schools are looking for workarounds. The president of Western Atlantic University School of Medicine suggested, “There are other ways you might say are proxies for race and ethnicity you can use to the maximum.” More workarounds over on StatNews.
An AI dictionary for all of us
Here’s an excellent glossary of AI from the VC firm, Andreessen. Every term you will ever hear is listed and defined in plain English. I hear these terms all the time, but never put in the sweat to look ‘em up. Check it out.
Happy fourth to all of your in The States. Thank you for reading.