Kickoff For September 22, 2025
I hope you're all taking time to take care of yourselves. Remember that, every so often it's OK to be selfish and put on your own oxygen mask (literally and figuratively speaking) before helping the person next to you with theirs.
With that out of the way, let's get Monday started with these links:
‘As thrilling as driving a sports car’: the Tokyo capsule tower that gave pod-living penthouse chic — A look at the (now-gone) Nakagin Capsule Tower in Tokyo and how it helped, at least for a short time, redefine living in a tiny space.
From the article:
Launched to critical acclaim, the Nakagin tower’s 140 capsules quickly sold out, and became highly sought after by well-heeled salarymen looking for a place to crash when they missed the last train home. Never intended to be full-time housing, the pods came stuffed with mod cons: en suite bathroom, foldout desk, telephone and Sony colour TV.
Scapegoating the Algorithm — Dan Williams argues that the problems many blame on social media and online disinformation, which are fuelled by the algorithms that drive them, go deeper than technology and have a longer history.
From the article:
[T]hese observations do not disprove the hypothesis that social media has exacerbated America’s epistemic challenges in recent years. Nevertheless, they serve to remind us that many of the problems attributed to social media can arise (and have historically arisen) in the absence of social media.
Humanlike? — Mike Dacey examines how we try to apply human emotions and qualities to animals, and how that doesn't capture the nuances of the differences (and similarities) between various living creatures.
From the article:
The errors we make in interpreting animal behaviour don’t always involve exaggerating intelligence – and in many cases, they do the opposite, causing us to overlook forms of intelligence that differ radically from our own. Consider the octopus: for a long time, we failed to recognise its cognitive sophistication, in part because of its alien biology – its neurons are distributed throughout its tentacles rather than centralised in a single brain.
Why AI chatbots lie to us) — Melanie Mitchell tackles the titular question, and explores why those lies continue to stream forth even as the models behind those AI chatbot supposedly improve.
From the article:
There is a simpler explanation for these behaviors. They are most likely a result of two factors: AI models’ pretraining, which induces them to “role-play,” and the special posttraining that these models receive from human feedback.