A Georgetown computer science professor arguing that attention is the scarce resource in knowledge work, followed by listener questions and a running check on what AI is actually doing to jobs.
The show is close to 400 episodes in and has changed shape more than once. The early run predates Slow Productivity and everything Newport now says about AI. Each episode opens with a self-contained deep dive on a single question, so the useful sort order is by question, not by number.
In order. If you only listen to one, make it the first.
Newport revisits the four rules of Deep Work on the book's tenth anniversary and says what he would change now. The additions are a hybrid attention model that syncs a team's remote days so nobody schedules meetings on them, and a rule against letting AI write for you.
His 1,600-person Digital Minimalism experiment split cleanly. The people who treated 30 days off as a detox relapsed immediately; the ones who filled the time with hobbies, reading and walking kept the change. He uses the basal ganglia to explain why waiting out a craving accomplishes nothing.
A Georgetown psychologist dismantles the popular idea that you have a daily stimulation budget with one line: people in Times Square would be comatose by noon. The real cost is context switching, and that reframe changes what you actually do about it.
Newport replaces the ladder-of-intelligence model of learning with a staircase, arguing the binding constraint is time rather than brainpower. The line the episode turns on: you can learn almost anything, but you cannot learn everything.
Brad Stulberg on why constant tracking makes people worse, including orthosomnia, the insomnia produced by chasing a perfect sleep score. Newport's addition is that measuring lets you simulate accomplishment without doing the difficult part.
Which is the real problem with "where do I start" lists. Each episode above links to its summary on this site, free and without an account, so you can find out which of the five is worth your evening before committing to any of them.
The AI Reality Check segments are 20 to 35 minutes on the week's AI news, and episodes titled 'Did the LLM Job Apocalypse Begin Last Week?' age exactly as fast as that suggests. On the full episodes the deep dive sits at the front; the listener questions in the back half are more variable.
The back-catalog problem is the same problem as the queue problem. Listening to summaries instead of full episodes and fitting long shows into a commute cover two other pieces of it.