Discussion about this post

User's avatar
James Ransom's avatar

Great piece, and it pairs nicely with Stefan Collini's recent LRB essay ('Squadrons of Pigs') on the dangers facing UK universities as institutions.

Two observations:

1. 'By thoughtfully adopting new technologies, universities can provide a temporal counterweight to short-run trends'. Your examples of this are aimed at individuals (on social media, citizen researchers), but there is an institutional version too. Business, local government, charities and healthcare providers all face uncertain futures, and academics can work with organisations outside the academy to shape how they take up the technology. Universities can be the spaces where discussions on AI and society take place (especially at the local level), but I'm not seeing this at present.

2. 'If AI continues to be able to execute more and more of the research process, we still need human experts with the breadth of knowledge to be able to check their work… the capacity to navigate from the medium to the long term successfully, on our terms, will require oversight and feedback from human experts'. How many experts do oversight and verification actually need? Perhaps a great many, to handle vibe-coded social science (especially if this emerges from, say, government departments alongside enthusiastic bloggers). Or perhaps there is a large fall in the number of human experts required – I speculate on what this could look like here: https://reskilled.substack.com/p/ai-labs-and-the-great-academic-migration

Hanno Hilbig's avatar

Great post Kevin! One thing I was wondering about: the claim that demand for expertise is going up makes sense, but AI itself is already pretty good at running the typical checks, at least for observational causal designs, which is what I mostly work on.

Assuming that most vibe-coded research is not super high quality, prob most of the demand for expertise can be supplied by AI itself. Of course for very good vibe-coded stuff, we may still need humans, but potentially fewer than are currently working in academia? Might be totally wrong here...

9 more comments...

Ready for more?