Note: This post was link-posted from the Asterisk by the Forum team, with the author's permission. The author may not see or respond to comments on this post. Summaries were auto-generated using Claude. Any mistakes are our own.
Church ladies. Y2K. Take any one of the following and show how it reflects the character of its author and of its age. Construction. Cotton-picking. Collaborative Registration Procedure eligibility. What does Louis XIV have in common with a chimpanzee? Rust belts. From Bauhaus to hacker house. Fully automated paranoid meltdowns. The future is coming whether you like it or not.
Asterisk is a quarterly journal of clear writing and clear thinking about things that matter (and, occasionally, things we just think are interesting). In this issue:
- The editors introduce Issue 15, which sets aside macro forecasts about automation to examine how particular jobs and institutions actually function and how AI may reshape them.
- Zilan Qian argues that China's apparent public optimism about AI reflects a learned survival response to disruption, formed by the mass state-sector layoffs of the 1990s, which means enthusiasm and fear are indistinguishable in the polls Western analysts cite.
- David Oks traces how India built its economy around exported IT and back-office labor, and argues AI threatens the labor side of that model even as the firms stay profitable, raising the question of what development path remains for poor countries when selling cheap remote labor stops working.
- Jesse Smith argues that the popular case for steering young people into the skilled trades oversimplifies the evidence, since college still carries a real earnings premium at a lower net cost than advertised, trades work brings serious injury and mortality risks, the best union jobs are far harder to access than assumed, and AI may erode trades work too rather than shielding it.
- Enlli McAleese explains how Zambia approved the HIV prevention drug lenacapavir in 12 days by relying on assessments already done by the EMA and WHO, showing that the gap in medicine access in low-income countries is largely a problem of institutional design rather than science or funding.
- Elizabeth Van Nostrand examines how pastors' wives perform extensive emotional and social labor that cannot be openly recognized as work, since acknowledging the effort would undermine the sincerity the role demands.
- Christine Kozobarich examines what the AI Village, a long-running experiment with dozens of AI agents sharing goals and tools, reveals about the collective failure modes of multi-agent systems, including lossy memory, runaway process over substance, and consensus overriding ground truth and human correction.
- I argue that work, not relationships, is what gives people dignity on impersonal terms, so a post-AI relational society would likely be more clannish and exclusionary than the one we have now.
- Kevin Munger argues that universities will endure by reinventing themselves around AI, and that academics should use the time AI frees up to read, mentor amateur researchers, and engage the public rather than simply publishing more.
- Alec Thompson traces how Scotland and several other countries replaced knowledge-rich curricula with skills-based teaching on weak theoretical grounds, and argues the reforms lowered attainment without reducing inequality, yet persist because they suit ideologies on both left and right.
- Celine Nguyen argues that taste is not an innate, irreplaceable human trait but a socially cultivated capacity, shaped through shared judgment and exchange rather than private intuition, and that Silicon Valley's framing of it as the last defensible edge against AI misreads both its history and its purpose.
- Abi Olvera describes how the U.S. diplomatic corps functions as a deliberately unglamorous, distributed fact-checking system whose structural incentives favour calibrated, continuously updated models of reality over the dramatic or advocacy-driven accounts that dominate public information.
- Anton Leicht argues that countries without domestic frontier AI face the greatest risk from the technology, absorbing its disruptions without capturing its gains, and that pursuing AI self-sufficiency would lock them into permanent geopolitical irrelevance rather than protect them.
- Dylan Matthews draws on the forgotten story of socialist inventors John and Mack Rust, whose mechanical cotton picker came with elaborate plans to share its gains but failed commercially, to argue that mission-driven firms like Anthropic face a hard limit: a company must win to wield influence, and winning tends to erode the ethical constraints that justified the mission.
Sign up for our newsletter to get Asterisk’s latest interviews, essays, and more.Huge thanks to everyone who helped bring Asterisk to life — we hope reading it brings you as much joy as making it did.
